Conversion of Sugarcane Trash to Nanocrystalline Cellulose and Its Life Cycle Assessment

: Sugarcane trash (SCT) is a promising, underutilized raw material for producing value-added bio-based materials. Nanocrystalline cellulose (NCC) production conditions were obtained from the experiment. On the other hand, bioethanol production conditions were retrieved from the secondary data. This study compared the environmental impact of SCT in NCC production to that of bioethanol. For NCC production, SCT was subjected to organosolv pretreatment (140, 160, or 180 °C) in a mixed solvent system (methyl isobutyl ketone (MIBK), ethanol, and water), bleached, and then hydrolyzed with different concentrations of sulfuric acid (50 and 58%) for varying times. Organosolv pretreatment at 180 °C removed 98.24 and 81.15% of the hemicellulose and lignin, respectively, resulting in 73.51 and 79.72% cellulose purity and recovery. In addition, bleaching increased the cellulose purity to 95.42%. Field Emission Transmission Electron Microscopy (FE-TEM) analysis showed that NCC's small 2:1 elliptical particles were found at the hydrolysis of 50% H 2 SO 4 for 45 min. The X-ray diffraction (XRD) pattern revealed 70% crystalline index values for NCC obtained from 50% H 2 SO 4 with 45 min retention times. Then, the optimum conditions of NCC production were used for LCA analysis (Sigmapro software). The analysis included global warming, marine ecotoxicity, fresh water, and human carcinogenic toxicity. NCC production's electricity consump-tion (freeze-dried step) was the highest environmental impact on LCA analysis.

NCC materials have recently attracted much attention because they are abundant, renewable, non-toxic, and biodegradable [13]. Moreover, such material can be applied in water and wastewater treatment [14], tissue engineering [15], and packaging [16]. As a result, the global demand for the nano-cellulose market size is foreseen to achieve USD 783 million by 2025 [17]. In addition, cane waste can be converted into bioethanol as a necessary renewable fuel to reduce carbon dioxide emissions [18][19][20][21]. In contrast, this work investigated the production of NCC, which began with biomass pretreatment using the acid catalyst and decomposed into cellulose, lignin, and hemicellulose [12]. Then, a life cycle assessment (LCA) is performed to determine NCC production using energy consumption and total products.

Effects of Reaction Temperature and Catalyst for Organosolv Pretreatment
The untreated sugarcane trash (SCT) was initially characterized to assess material conditions before undergoing the organosolv pretreatment [22]. The results of the characterization and organosolv pretreatment are shown in Table 1. It was found that the mean (±standard deviation) percentages of cellulose, hemicellulose, lignin, and ash were 33.35 ± 0.2, 20.26 ± 0.4, 22.70 ± 0.8, and 7.13 ± 0.1%, respectively, which were similar to those reported by Powar et al. [10] and Cardoen et al. [11]. After pretreatment and bleaching (Table 1), 76.22% cellulose in the pretreated, bleached SCT made it a potential NCC production source. The organosolv pretreatment of SCT was conducted using 25, 42, and 33 v/v% of methyl isobutyl ketone (MIBK), ethanol, and water, respectively, with 0.05 M sulfuric acid as the catalyst. The reaction contained 100 mL of working volume, with the pretreatment temperature varying from 140 to 180 °C for 40 min. H2SO4 was chosen as the catalyst in most clean fractionation processes reported due to its cost-effectiveness. In addition, this acid catalyst could increase the solubilization of hemicellulose and lignin [23]. Therefore, MIBK was selected as the solvent and washing agent since this reagent has high lignin solubility (more than 90%) and results in the enrichment of cellulose in the solid fraction [23].
The pretreatment temperature significantly influenced the delignification. When the pretreatment temperature increased from 140 to 180 °C, the lignin removal improved from 66.99 to 81.15%. A higher pretreatment temperature could provide more energy to facilitate the cleavage of lignin bonds, resulting in higher lignin removal [24]. Similarly, these trends occurred for the hemicellulose composition. As the temperature rose from 160 to 180 °C, the hemicellulose removal efficiency increased from 83.83% to 98.24%. Removing hemicelluloses involves the cleavage of covalent bonds that link hemicelluloses to lignin [25].
On the other hand, the cellulose recovery declined gradually to 79.72%, from 97.56% when the temperature rose from 140 to 180 °C. The reduction of cellulose recovery because the cleavage of the glycosidic bonds influences the declining cellulose recovery and the hydrolysis of cellulose to glucose and cello-oligosaccharides [26]. Conversely, the percentage of cellulose purity increased gradually from 66.60 to 73.51% when the temperature was enhanced from 140 to 180 °C. Out of these results, the optimum pretreatment temperature at 180 °C was fixed for a further experiment, which recorded the lignin composition at 11.01%, with 79.92% cellulose recovery and 98.24% hemicellulose removal efficiency. The temperature of 180 °C was selected because it produced the highest cellulose purity (73.51%). Figure 1 shows the appearance of the SCT samples, which became darker compared to untreated SCT as the temperature increased (Figure 1a,d).

Bleaching
The bleaching step aims to raise cellulose purity after the cane trash undergoes organosolv pretreatment [27]. The H2O2 in the base conditions was used for this experiment. The lignin removal increased from 81.15 to 98.48% (Table 1), and cellulose purity rose sharply from 73.51 to 95.42%. Conversely, the amount of cellulose recovery was substantially reduced to 27.94%. The increased lignin removal shows the effectiveness of the bleaching reaction. However, the radicals generated may randomly cause alkaline-catalyzed cleavages of polysaccharide chains [28], leading to cellulose degradation [29]. The appearance of pretreated and bleached SCT (PB-SCT) seems to be whiter than the pretreated SCT ( Figure 1). The color change might imply increased and decreased cellulose and lignin content [30].

Morphological Analysis
Hydrolysis of the PB-SCT was studied using 50% and 58% sulfuric acid at 45 °C for 30 min, 45 min, and 60 min. The hydrolysate became a darker brown after 30 min of hydrolysis using 58% sulfuric. This dark color may have resulted from side reactions, such as dehydration [31,32]. Figure 2 shows the particle size of the hydrolysis product. The average particle size was estimated to visualize the trend of the changes in particle sizes.
The obtained NCC seemed to have a 2:1 elliptical shape, with an average of 96 nm length and an average of 45 nm diameter for the NCC from hydrolysis reactions with 50% sulfuric acid for 45 min. The size of NCC was smaller at a longer hydrolysis time of 60 min, and NCC became spherical with an average diameter of 38 nm. Acid hydrolysis during the longer reaction time eliminates the amorphous regions and portions of the crystalline component and reduces the crystallite sizes [33,34]. Various diameters of spherical NCC from agricultural waste were produced, including 95.9 nm [35], 30-40 nm [36], 42-82 nm [37], and 33-67 nm [38]. Interestingly, the coagulation between particles was found even though samples were split by suspending them in water before being pipetted onto the FE-TEM grid [37]. Expectedly, the impurity of a sulfated group (S−O2) from acid hydrolysis might cause this coagulation. However, sulfuric acid was chosen for hydrolysis since it can produce smaller NCC [38] and also can effectively hydrolyze amorphous regions completely [39] with a higher crystallinity index [33]. In addition, sonication was carried out for 10 min at the end of hydrolysis in ice to decrease the nanocellulose particle size and avoid overheating. On the other hand, sonication longer than 10 min can cause some degradation of the NCC [40].

X-ray Diffraction Analysis (XRD)
The XRD patterns of samples subjected to 50% H2SO4 for 45 min and 60 min hydrolysis are presented in Figure 3. The diffractograms showed two diffraction angles at 2θ of about 15° and 22°, indicating the typical native cellulose structure (cellulose I) [41,42]. In addition, diffraction angles at around 18° and 22° were noted as the NCC's amorphous and crystal regions [42]. The peak at approximately 2θ = 22° appeared sharper and more substantial and indicated the crystalline domain [43]. However, the border crystal peak was also found [44]. The XRD pattern showed that at 2θ = 22°, there were intensities of 15,134 and 9420. At the same time, the levels were 4533 and 3455 for 2θ = 18° for 50% H2SO4 at 45 and 60 min, respectively. Of these 2θ intensities, the crystallinity indexes of NCC with 50% H2SO4 for 45 and 60 min were 70% and 63%, respectively. However, the changing percentage of the crystallinity index was associated with the hydrolysis condition. Kargarzadeh et al. [43] reported that a further increase in the hydrolysis time could lead to a decrease in the NCC shape and a decline in the NCC crystallinity.  Figure 4 shows the FTIR spectra of NCC after undergoing hydrolysis with 50% H2SO4 for 45 mins and 60 mins, respectively. Overall, both samples exhibit almost identical spectra, indicating no changes in their chemical composition during sulfuric acid hydrolysis [45]. The absorbance peaks in the 3500-3200 cm −1 are associated with the vibration of a hydroxyl group (-OH) stretching of cellulose due to intra-and intermolecular hydrogen bonds. In addition, the peaks around 2900-2800 cm −1 can correspond with C-H stretching vibration in cellulose and hemicellulose. The attribution to the C=O stretching of uranic esters and acetyl groups of the hemicellulose occurred at 1733 cm −1 . The exact peak also represented ester linkages of the carboxylic group in hemicellulose and lignin [45]. The prominent band at 1640 cm −1 was attributed to the O-H of the absorbed water [42,46]. The peaks at 1428 and 1370 cm −1 correlated with the -CH2 scissoring motion of cellulose and C-H bending, respectively [47]. The carbonyl group's vibration peaked at 1322 cm −1 [45]. The existence of a sulfated group (S-O2) at 1160 cm −1 was referred to as sulfuric acid from cellulose hydrolysis [48]. A peak at 1111 cm −1 reflected glucose ring stretching [49,50], 1060 cm −1 with the vibration of C-O-C in the pyranose ring [35], and 897 cm −1 with C-O-C extension at β-glycosidic linkages between glucose [51].

Life Cycle Impact Assessment
2.6.1. Global Warming Impact on NCC Production Figure 5 shows that 174.6 kg of CO2 eq was produced from 1 kg of SCT or for the production of 0.011 kg of NCC, most of which was contributed by power and MIBK with 136.6 and 31.4 kg CO2 eq, respectively. The high impact of electricity contribution to global warming was related to the high electricity usage (667.6 MJ) and electricity generation, in which 90% of electricity generation is produced from fossil fuel [52]. In addition, fossil fuel contributes to global warming due to the carbon dioxide (CO2) released during fossilfuel combustion [53]. The enormous MIBK contribution relates to the high chemical quantity. On the other hand, MIBK is a volatile organic compound that is well known as one of the significant contributors to air pollution [54] and global warming impact [55]. The other remaining consequences, which need more attention, are freshwater and marine ecotoxicity because of the high amount of wastewater ( Table 2 ) during NCC production.

Marine Ecotoxicity Impact of NCC Production
Generally, Figure 6 depicts that 13.0 kg of 1,4-DCB contributed to the marine ecotoxicity impact by NCC production, most of which was due to power (11.3 kg 1,4-DCB) and MIBK (1.2 kg 1,4 DCB). The high contribution of electricity toward marine ecotoxicity was related to the metals generated (such as Se, Ni, and Cu) during fossil fuel burning, in which most of these metals will end up in the water and poison the marine ecosystem [56]. The high impact of MIBK on marine ecotoxicity causes massive usage. Although MIBK has low toxicity toward microorganisms and aquatic organisms [57], Chipman [58] reported that inappropriate disposal in large quantities could cause toxicity to organisms in the environment.

Freshwater Ecotoxicity Impact of NCC Production
Broadly, Figure 7 explains that in producing 0.011 kg of NCC, 10.4 kg of 1,4-DCB was exposed to freshwater damage. MIBK and power made more substantial contributions of 1.0 and 8.9 kg 1,4-DCB, respectively. The considerable freshwater impact from electricity was caused by the high number of metals (such as Cu, Se, and Ba) generated during fossil fuel combustion that is consequently in water [56] as pollutants. Coal constitutes the second-largest share of the electricity output (22%) in Thailand, and according to Atilgan and Azapagic [59], coal power is associated with ecotoxicities (marine and freshwater). MIBK is considered a volatile organic compound [54] that can interact with the other pollutants (NOx and SOx) in the atmosphere and atmospheric water, resulting in dry deposition and wet deposition [60]. These depositions then flow and end up in the water catchment, polluting the freshwater ecosystem. The last impact that needs to be assessed is human carcinogenic toxicity.  Figure 8 shows that 7.4 kg of 1,4-DCB was produced during the production of 0.011 kg of NCC, with power making the most significant contribution, followed by MIBK of 5.5 and 0.8 kg of 1,4-DCB, respectively. The considerable electricity impact is rendered by pollutants emitted (SOx and NOx) from fossil fuel combustion [61]. As a result, the population can be exposed to these contaminants that are potentially carcinogenic for humans through ingestion, inhalation, and dermal contact [62]. MIBK's more considerable contribution is that this chemical evaporates quickly at room temperature. Consequently, inhalation exposes people to this solvent, causing nose and lung irritation [63]. Power invariably causes the most significant environmental impact of these four categories (global warming, marine ecotoxicity, freshwater ecotoxicity, and human carcinogenic toxicity). All the production steps for NCC (Table 2) require energy. However, the freeze-dried process consumes the highest power of 651.92 MJ among those steps. The four impact analyses show that NCC production produces high environmental impacts (174.6 kg CO2 eq, 13.0 kg 1,4-DCB, 10.4 kg 1,4-DCB, and 7.4 kg 1,4-DCB for global warming, marine ecotoxicity, freshwater ecotoxicity, and human carcinogenic toxicity impact, respectively).

Materials
The dried SCT, supplied by a sugar mill in Thailand, was milled and sieved to 0.5-1.0 mm. The chemical composition was analyzed using the NREL method [64], which is based on the acid hydrolysis principle in which 0.3 g of the sample was hydrolyzed using 3 mL of 72% sulfuric acid for 2 h with vortex mixing every 15 min. Then, 84 mL of deionized water was added to the solution immediately, followed by autoclaving at 121 °C for 1 h. Cellulose and hemicellulose were determined from the liquid fraction, which had been neutralized using CaCO3 that was analyzed using HPLC (Shimadzu, Kyoto, Japan) equipped with an Aminex HPX-87H (Biorad, Hercules, CA, USA) column and a refractive index detector (RID). Lignin was determined from the solid heating fraction at 105 °C for 4 h, then burnt at 575 °C for 2 h [65].

Pretreatment Using Acid Catalyst
The biomass (10% w/v) was pretreated in a ternary solvent comprising MIBK, ethanol, and water (25,42, and 33% v/v, respectively) with 0.05 M H2SO4 as a catalyst for 140, 160, and 180 °C for 40 min with an initial pressure of nitrogen at 20 bars and stirring at 100 rpm in a 1 L high-pressure batch reactor (Parr Instruments, Moline, IL, USA) [23].
First, the reaction was stopped by quenching in an ice bath. Then, the slurry was passed through 20-25 μm filter paper. Finally, the residue was immersed in 30 mL of MIBK, filtered, neutralized with DI water, and dried at 70 °C in a hot-air oven. The pretreated SCT pulp yield was calculated as follows: where the pulp yield is the percentage of the remaining sample compared to the initial weight (%), and f and i are the weights of the remaining solid samples after the organosolv treatment and the initial raw material, respectively. The biomass compositions were determined based on the remaining contents in solid residues compared with the respective contents in the native biomass. The liquid fraction of the aqueous-organic mixture was separated into two phases by adding water in a separator funnel. Afterwards, the mixture was left at room temperature for 20 min for complete phase separation. The organic phase, which contained lignin and MIBK (the major component), was evaporated at 70 °C to recover lignin.

Extraction
The PB-SCT were extracted using 50% and 58% sulfuric acid (1:25 w/v) at 45 °C and 350 rpm for 30, 45, and 60 min [27]. A ten-fold amount of cold DI water was added to the suspension. The mixture was centrifuged at 8000 rpm for 20 min and dialyzed against DI water. The solid fraction was recovered using centrifugation at 8000 rpm for 60 min. Finally, the particles were treated using sonication at 20 kHz for 10 min with 40% amplitude. Some NCC suspension was separated for particle analysis, and another suspension sample was freeze-dried (FD) to remove the water content [66].

Characterization Procedures
The total yield of the NCC synthesis was calculated as a percentage ratio of the freeze-dried NCC based on the initial weight of cellulose in the raw material. Several instruments were used to analyze the physicochemical properties: an FE-TEM, a Fouriertransform infrared spectroscope (FTIR), and an X-ray diffractometer (XRD).
The FE-TEM preparation was conducted by diluting 0.1-0.3% w/w of NCC in water. Then, one drop of each sample was deposited on a formvar or carbon-coated grid. After 1 min, the water drop was removed by dabbing it with tissue paper [67]. The FE-SEM micrographs were recorded at 40,000× magnification. The ImageJ unit estimated the sample diameter [37]. X-ray diffraction patterns were obtained using a D/max-2500 X-ray diffractometer (RigakuDenki, Tokyo, Japan) using Cu-K radiation (0.154 nm) in 2 θ = 5-40° at a scan rate of 2°/min. The crystallinity indices (I ) were counted based on the Segal method [68]: where I002 and Iam are the peak intensities of the crystalline and amorphous materials, respectively, FT-IR was used to study the presence of chemical groups. Each sample was mixed with KBr. Their spectra were recorded at a resolution of 2 cm −1 with a spectral range of 4000-450 cm −1 [36].

Life Cycle Assessment
The NCC production was assessed for its environmental impact based on the product life cycle. The evaluation commenced by determining the energy consumption, raw material input, and output per functional unit using the SimaPro Software. Afterwards, the life cycle impact assessment procedure, based on the Ecoinvent database, was applied to calculate the environmental impact according to ReCiPe 2016 Midpoint/Endpoint (H) V1.04/World (2010).

Goal and Scope Definition
The LCA study evaluated the environmental impact assessment of NCC production from sugar cane trash (SCT). The functional unit of cellulose extraction was 1 kg of SCT for NCC production. A total of 1 kg of SCT was converted to 0.011 kg NCC. Gate-to-gate was selected for the system boundary of these experiments. The system began with SCT grinding and organosolv pretreatment (using an H2SO4 catalyst) to obtain cellulose and lignin, followed by bleaching, extraction, sonication, and freeze-drying.

Life Cycle Inventory
The life cycle inventory data, including each stage of the NCC production process, are exhibited in Table 2. In addition, data regarding water consumption, electricity, and reagents were measured in the laboratory.

Conclusions
In this study, we determined the environmental impact of two processes using SCT as a raw material. At first, SCT was pretreated by organosolv using an acid catalyst and then used in NCC production. Next, the optimum conditions of NCC production were obtained by an experiment consisting of three steps: bleaching, extraction and sonication, and freeze-drying. Finally, the LCA of the process was analyzed. It was found that NCC production had the most significant severe environmental damage from electricity used in freeze-drying. This finding is essential for the future development of NCC production.