Ultrasonic Irradiation Coupled with Microwave Treatment for Eco-friendly Process of Isolating Bacterial Cellulose Nanocrystals

The isolation of crystalline regions from fibers cellulose via the hydrolysis route generally requires corrosive chemicals, high-energy demands, and long reaction times, resulting in high economic costs and environmental impact. From this basis, this work seeks to develop environment-friendly processes for the production of Bacterial Cellulose Nanocrystals (BC-NC). To overcome the aforementioned issues, this study proposes a fast, highly-efficient and eco-friendly method for the isolation of cellulose nanocrystals from Bacterial Cellulose, BC. A two-step processes is considered: (1) partial depolymerization of Bacterial Cellulose (DP-BC) under ultrasonic conditions; (2) extraction of crystalline regions (BC-NC) by treatment with diluted HCl catalyzed by metal chlorides (MnCl2 and FeCl3.6H2O) under microwave irradiation. The effect of ultrasonic time and reactant and catalyst concentrations on the index crystallinity (CrI), chemical structure, thermal properties, and surface morphology of DP-BC and BC-NC were evaluated. The results indicated that the ultrasonic treatment induced depolymerization of BC characterized by an increase of the CrI. The microwave assisted by MnCl2-catalyzed mild acid hydrolysis enhanced the removal of the amorphous regions, yielding BC-NC. A chemical structure analysis demonstrated that the chemical structures of DP-BC and BC-NC remained unchanged after the ultrasonic treatment and MnCl2-catalyzed acid hydrolysis process.


Introduction
The most advantageous characteristics of the bio-based edible film are their edibility and inherent biodegradability [1]. Various biopolymers have been explored to reduce the use of non-degradable petroleum-based materials such as cellulose, chitosan, starch, collagen, pectin, etc. [2]. However, problems of strong hydrophilic character, high degradation, and inadequate mechanical properties in moist environments still limit the applications of biopolymers [3,4]. To become more applicable in practice, biopolymers have to be modified in terms of properties and functionalities [5]. In food packaging applications, for instance, the incorporation of reinforcement fillers [6,7] into the crystallites [41]. It generates a low-density surface charge with limited NCC dispersibility, which tends to promote flocculation in aqueous suspensions [42]. HCl is less corrosive than H 2 SO 4 , and though the yield is lower [41], it permits a significant increase in thermal stability of NCC [43]. To reach high yield value, a highly-concentrated aqueous solution of HCl is needed under hydrothermal conditions at 110 • C for a long period of the reaction [35].
As described in the previous paragraph, high-yield production of NCC is obtained using an excessive amount of mineral acids. Pollution to the environment, corrosion to the equipment, and the difficulty of controlling the reaction are the major limitations to synthesis using acid hydrolysis [44]. To overcome these issues, the aim of this study is to develop a fast, highly-efficient, and eco-friendly preparation method for the extraction of cellulose nanocrystals from Bacterial Cellulose, BC. A two-step process is considered, yielding Bacterial Cellulose NanoCrystals, BC-NC, namely: (1) partial depolymerization of BC under ultrasonic irradiation, (2) extraction of crystalline regions using microwave assisted by MnCl 2 -catalyzed hydrolysis. The effect of irradiation time on the partial depolymerization process and impact of MnCl 2 concentration during the hydrolysis treatment is evaluated on the chemical structure, crystallinity index, thermal properties, and surface morphology of irradiated Depolymerized Bacterial Cellulose DP-BC and extracted BC-NC

Materials
Nata de coco (BC pellicles) was collected from a local market, in Cilegon (Banten, Indonesia) region. Sodium hydroxide, NaOH, hydrochloric acid, HCl, and ethanol, C 2 H 5 OH were obtained from Thermo Fischer Scientific. Metal chlorides MnCl 2 , FeCl 3 .6H 2 O were purchased from Merck Indonesia (Jakarta, Indonesia). Commercial Microcrystal Cellulose, MCC was purchased from Sigma-Aldrich (Saint-Quentin, France) and Commercial Nanocrystal Cellulose, NCC from CelluForce, QC, Canada. All the reagents and chemicals are used as a laboratory grade without further purification. Demineralized water (conductivity of 0.06 mScm −1 ) produced by a purification chain was used for all experiments.

BC Preparation
BC pellicles were maintained in 0.5% NaOH (w/v) at room temperature for 24 h, followed by rinsing in the drained water until a neutral pH was attained; any chemicals used in the nata de coco production removed. The cellulose was then sun dried for two days, powdered, and sieved through a 149 µm sieve (100 Mesh).

Partial Depolymerization of BC
The ultrasonic irradiation was carried out using an ultrasonic processor (Vibra Cell, Type 72434, 100 Watts, horn diameter: 1.0 mm, Fisher Scientific, Illkirch, France). One half of a gram (1% w/v) of BC powder was introduced into a 100-mL flat-bottom flask with a mixture of water/ethanol (50% w/w). The ultrasonic horn was placed at the center of the suspension, while the temperature was maintained at room temperature with a circulating water condenser. The suspension was constantly stirred at 300 rpm using magnetic stirring bar. All sonication runs were carried out at 20 kHz by varying irradiation times of 10, 20, 30, 60, and 120 min. After that, each sample of DP-BC was immediately washed with water and filtered with Whatman filter paper no. 1 until the filtrate was neutral. The DP-BC was then oven dried at 70 • C for 24 h.

Extraction of Crystalline Regions
After ultrasonic irradiation treatment, the DP-BC (oven dry 0.3 g) was put into a microwave reaction vial (30 mL, G30), and 15 mL water of 0.1 mol/L HCl with a concentration of MnCl 2 in the range 0; 1; 2.5; 5% w/v was added. The suspension was then placed into a microwave reactor (Anton Paar, Monowave 300) and heated as follows: (1) heating to 125 • C in 3 min; (2) maintaining the temperature at 125 • C for 30 min; and (3) cooling to 30 • C within 7 min. During the protocol, the suspension was stirred with a magnetic stirring bar at 1200 rpm. Upon completion of the hydrolysis, the vial was removed from the microwave oven and cooled at ambient temperature. The BC-NC suspension was transferred to a 50 mL plastic centrifuge tube and centrifuged at 12,000 rpm for 5 min (Jouan, MR 1812 Refrigerated Centrifuge, MN, USA) to remove residual acid and chemicals. The precipitate was purified by five washing cycles with deionized water followed by centrifugation at 12,000 rpm for 5 min. The BC-NC was then oven dried at 70 • C for 24 h before characterizations.

Fourier Transform Infrared (FT-IR)
FT-IR study was conducted to determine the functional groups present in the cellulose. The measurements were performed using a Nicolet iS5 spectrometer (Thermo Scientific, Whatman, MA, USA). Spectra were obtained between 4000 and 400 cm −1 at a resolution of 4 cm −1 and scanning speed of 20 mm/sat.

X-Ray Diffraction (XRD)
XRD patterns of the cellulose were performed in a D8 Advance (Bruker, Bremen, Germany). Samples were examined with a scanning angle of 2θ from 10 • to 40 • at a rate of 1 • /min with the CuKα filtered radiation. The crystallinity index, CrI was calculated from the diffraction intensity data using deconvolution method [45]. In which the diffraction profile was fitted by Gaussian function to find the contribution of each individual peak relative to the crystallographic planes and the amorphous background. The CrI was calculated Hermans equation as follows: where A am is the amorphous area, and A cr is the sum of the area of the 101, 10ī, 002, 040 peaks.

Differential Scanning Calorimetry (DSC)
DSC was carried out to analyze the thermal behavior of the cellulose. The samples were characterized on a DSC Q100 (TA Instruments, DE, USA) under constant nitrogen flow (50 mL/min), from 25 to 400 • C, at a heating rate of 10 • C/min.

Transmission Electron Microscopy (TEM) and Scanning Transmission Electron Microscopy STEM)
The morphology of BC-NC suspension was measured by using the high-resolution JEOL-2100F TEM (Jeol, Akishima, Tokyo, Japan) in TEM and STEM mode. Samples were conventionally deposited on carbon coated copper grids and a negative staining was achieved using uranyLess solution (Delta Microscopies, Toulouse, France). The size and diameter distribution particle were measured by ImageJ (version 1.41 h) and origin pro-8 software.

Chemical Structure
FT-IR spectroscopy was used to investigate changes in the chemical structure of cellulose sample before and after the treatments. The spectra displayed the intensity of absorption of the functional groups between 4000 and 400 cm −1 , which can identify the chemical bond in the cellulose molecule. The absorption bands for characteristic chemical groups of the raw material (native BC) and the treated celluloses (DP-BC and BC-NC) can be observed in Figure 1, and the typical vibration bands are listed in Table 1. The FT-IR spectra were divided into two parts: (1) H-bonding region from 4000 to 2600 cm −1 , and (2) fingerprint region from 1800 to 400 cm −1 [46]. The broad peak in the 3650-3000 cm −1 bands was assigned to O-H stretching vibrations, which are characteristic of the hydroxyl groups generally present in cellulose, water, and lignin. In this region, intramolecular hydrogen bonds appeared at 3342 cm −1 and 3432 cm −1 , and were attributed respectively to the two crystalline cellulose allomorphs, cellulose Iα and cellulose Iβ [47]. groups generally present in cellulose, water, and lignin. In this region, intramolecular hydrogen bonds appeared at 3342 cm −1 and 3432 cm −1 , and were attributed respectively to the two crystalline cellulose allomorphs, cellulose I and cellulose I [47].   [48], these hydroxyl groups were responsible for the stiffness in the polymer chain, and for allowing the linear polymers to form sheet structures. The strong vibration band around 2895 cm −1 corresponded to C-H stretching vibrations [49]. This band may be associated with a hydrocarbonate linear chain. Higher values in this specific band are correlated to a decrease in the calculated total crystallinity value [50]. An intense band at 1429 cm −1 band can be assigned to the bending of asymmetric angular deformation of C-H bonds. The band found between 1420 to 1430 cm −1 was associated with the amount of the cellulose ordered form, while the band appearing at 898 cm −1 was assigned to the disordered region [51]. The 1163 cm −1 band was assigned to asymmetrical stretching of C-O-C glycoside bonds. According to Börjesson and Westman [48], these hydroxyl groups were responsible for the stiffness in the polymer chain, and for allowing the linear polymers to form sheet structures. The strong vibration band around 2895 cm −1 corresponded to C-H stretching vibrations [49]. This band may be associated with a hydrocarbonate linear chain. Higher values in this specific band are correlated to a decrease in the calculated total crystallinity value [50]. An intense band at 1429 cm −1 band can be assigned to the bending of asymmetric angular deformation of C-H bonds. The band found between 1420 to 1430 cm −1 was associated with the amount of the cellulose ordered form, while the band appearing at 898 cm −1 was assigned to the disordered region [51]. The 1163 cm −1 band was assigned to asymmetrical stretching of C-O-C glycoside bonds.

Crystallinity Index
The crystallinity index of the cellulose was analyzed by X-ray Diffraction analysis. The CrI was calculated by curve-fitting process where individual crystalline peaks were extracted from the diffraction intensity profiles [52,53]. X-ray diffractogram of BC sample was fitted by Gauss function; the results are shown in Figure 2a, and the optimum results of the partial depolymerization step and the extraction of crystalline domains step are presented in Figure 2b. The integrated peak area obtained of each fitting curve is shown in Table 2. There were at least eight peaks that had been separated from the diffraction intensity profiles but only four distinct characteristic peaks at 2 = 14.6°, 16.8°, 22.6°, and 34.1°, which were considered to correspond to 101, 10ī, 002, and 040 crystallographic planes [54,55]. The broad peaks were attributed to the amorphous contribution. The assumption for this analysis was that the amorphous contribution increase was the main contributor to peak broadening. According to Park et al. [45], the other intrinsic factors that influence peak broadening were crystallite size and non-uniform strain within the crystal. These assumptions were then utilized to carry out an investigation of the crystallinity for all the samples. After subtracting the amorphous regions from the whole samples, the CrI was calculated by dividing the remaining diffractogram area due to crystalline cellulose by the total area of the original diffractogram. The samples presented correspond to the optimum results of the partial depolymerization step (DP-BC) which have been further hydrolyzed into the microwave reactor (BC-NC).

Thermal Properties
Crystallinity is an important parameter, which can greatly affect the physical properties of biodegradable polymers. The identification of neat polymers, copolymers, polymer blends, and composites, as well as the determination of their purity and stability, are generally described by DSC. Amorphous polymers exhibit a glass transition temperature while crystalline or semi-crystalline polymers may possess glass transition temperature, a freezing and melting temperature with various freezing, and melting enthalpies [56]. In this work, the glass transition temperature, Tg, melting point, Tm, and decomposition temperature, Td were investigated to interpret thermal behaviors of the extracted celluloses. The DSC thermograms for the BC, DP-BC, BC-NC and the reference materials, commercial Microcrystal Cellulose, MCC and Nanocrystal Cellulose, NCC were registered at a heating rate of 10 °C/min and depicted in Figure 3. The thermograms revealed that BC exhibited different pattern from DP-BC and BC-NC. The heat-flow curve of BC displayed a small inflection of the baseline around 105-110 °C, which is the glass transition temperature, Tg, and is followed by an  There were at least eight peaks that had been separated from the diffraction intensity profiles but only four distinct characteristic peaks at 2θ = 14.6 • , 16.8 • , 22.6 • , and 34.1 • , which were considered to correspond to 101, 10ī, 002, and 040 crystallographic planes [54,55]. The broad peaks were attributed to the amorphous contribution. The assumption for this analysis was that the amorphous contribution increase was the main contributor to peak broadening. According to Park et al. [45], the other intrinsic factors that influence peak broadening were crystallite size and non-uniform strain within the crystal. These assumptions were then utilized to carry out an investigation of the crystallinity for all the samples. After subtracting the amorphous regions from the whole samples, the CrI was calculated by dividing the remaining diffractogram area due to crystalline cellulose by the total area of the original diffractogram. The samples presented correspond to the optimum results of the partial depolymerization step (DP-BC) which have been further hydrolyzed into the microwave reactor (BC-NC).

Thermal Properties
Crystallinity is an important parameter, which can greatly affect the physical properties of biodegradable polymers. The identification of neat polymers, copolymers, polymer blends, and composites, as well as the determination of their purity and stability, are generally described by DSC. Amorphous polymers exhibit a glass transition temperature while crystalline or semi-crystalline polymers may possess glass transition temperature, a freezing and melting temperature with various freezing, and melting enthalpies [56]. In this work, the glass transition temperature, T g , melting point, T m , and decomposition temperature, T d were investigated to interpret thermal behaviors of the extracted celluloses. The DSC thermograms for the BC, DP-BC, BC-NC and the reference materials, commercial Microcrystal Cellulose, MCC and Nanocrystal Cellulose, NCC were registered at a heating rate of 10 • C/min and depicted in Figure 3. The thermograms revealed that BC exhibited different pattern from DP-BC and BC-NC. The heat-flow curve of BC displayed a small inflection of the baseline around 105-110 • C, which is the glass transition temperature, T g , and is followed by an endothermic peak with the onset, T m = 113.8 • C. In the treated celluloses, DP-BC and BC-NC, with the increased in crystalline content (CrI), the endothermic peak shifted toward a higher temperature. This slightly marked peak could be attributed to the presence of an amorphous region [57]. The curve of DP-BC showed an endothermic peak at 310-370 • C, which appeared to be a melting temperature at T m = 348.7 • C. The peak was followed by a degradation temperature at T d = 381.3 • C of the cellulosic material. A similar result was found for BC-NC: an endothermic peak was detected around 260-290 • C with the onset temperature at T m = 282.8 • C then followed by degradation at T d = 318.6 • C. The T g of both samples, DP-BC and BC-NC disappeared or became difficult to detect because of the partial removal of the amorphous regions. In the treated celluloses, DP-BC and BC-NC, with the increased in crystalline content (CrI), the endothermic peak shifted toward a higher temperature. This slightly marked peak could be attributed to the presence of an amorphous region [57]. The curve of DP-BC showed an endothermic peak at 310-370 °C, which appeared to be a melting temperature at Tm = 348.7 °C. The peak was followed by a degradation temperature at Td = 381.3 °C of the cellulosic material. A similar result was found for BC-NC: an endothermic peak was detected around 260-290 °C with the onset temperature at Tm = 282.8 °C then followed by degradation at Td = 318.6 °C. The Tg of both samples, DP-BC and BC-NC disappeared or became difficult to detect because of the partial removal of the amorphous regions.

Morphology
The morphology of nanocrystals after the hydrolysis treatment was characterized by TEM and STEM observations. Figure 4 presents a comparison of TEM and STEM micrograph of sample BC-NC obtained from a 30 min microwave reaction with MnCl2 catalyst (5% w/w) at 125°C to commercial NCC. Both samples were prepared in the same conditions, and as shown, are similarly constituted of a mixture of fibrillated structure with variable length and smaller needles or nano-rods. It seems that the fibrillated structures are in fact constituted of the densely-packed needles. The BC-NC nano-rods are of (164.51 ± 7.56) nm in length with an average diameter of (25.05 ± 2.80) nm. They look like the commercial forms which are (90.94 ± 10.05) nm in length with an average diameter of (12.58 ± 0.87) nm. This result confirmed the crystalline structure already described with a very close similarity to commercially-available nanocrystalline cellulose.

Morphology
The morphology of nanocrystals after the hydrolysis treatment was characterized by TEM and STEM observations. Figure 4 presents a comparison of TEM and STEM micrograph of sample BC-NC obtained from a 30 min microwave reaction with MnCl 2 catalyst (5% w/w) at 125 • C to commercial NCC. Both samples were prepared in the same conditions, and as shown, are similarly constituted of a mixture of fibrillated structure with variable length and smaller needles or nano-rods. It seems that the fibrillated structures are in fact constituted of the densely-packed needles. The BC-NC nano-rods are of (164.51 ± 7.56) nm in length with an average diameter of (25.05 ± 2.80) nm. They look like the commercial forms which are (90.94 ± 10.05) nm in length with an average diameter of (12.58 ± 0.87) nm. This result confirmed the crystalline structure already described with a very close similarity to commercially-available nanocrystalline cellulose.

Effect of Ultrasonic Irradiation on The Depolymerization Cellulose
The term "ultrasonic" describes sound waves with a frequency greater than 20 kHz. Many studies have reported the exposure to this wave is responsible for a number of physical and chemical changes. The ultrasonication was the adopted method here to carry out partial depolymerization of native BC into microfibers. The utilization of ultrasonic waves offers a simple and versatile tool for synthesizing micro or nanostructured materials that are often unavailable by conventional methods. In this work, native BC was irradiated into a mixture of water/ethanol (50% w/w) at constant power of 100 W and frequency of 20 kHz by varying irradiation times. The influence of the length of ultrasonic irradiation period on the CrI is presented in Table 3. From the table, it could be concluded that the increase of CrI is dependent on the irradiation time. Ultrasonic irradiation in water/ethanol induced partial depolymerization of BC with a CrI increase of 8.4% during the first 30 min. The maximum CrI was observed to be 71.4% at 120 min. The results indicated that the irradiation leads to the rupture of amorphous cellulose chains. The disintegration of amorphous regions may be explained by acoustic cavitation. As native BC in a liquid medium was exposed to ultrasonic irradiation, the acoustic waves induce alternating high and low pressure; this creates bubbles (i.e., cavities) and makes them oscillate. A bubble can grow while

Effect of Ultrasonic Irradiation on The Depolymerization Cellulose
The term "ultrasonic" describes sound waves with a frequency greater than 20 kHz. Many studies have reported the exposure to this wave is responsible for a number of physical and chemical changes. The ultrasonication was the adopted method here to carry out partial depolymerization of native BC into microfibers. The utilization of ultrasonic waves offers a simple and versatile tool for synthesizing micro or nanostructured materials that are often unavailable by conventional methods. In this work, native BC was irradiated into a mixture of water/ethanol (50% w/w) at constant power of 100 W and frequency of 20 kHz by varying irradiation times. The influence of the length of ultrasonic irradiation period on the CrI is presented in Table 3. From the table, it could be concluded that the increase of CrI is dependent on the irradiation time. Ultrasonic irradiation in water/ethanol induced partial depolymerization of BC with a CrI increase of 8.4% during the first 30 min. The maximum CrI was observed to be 71.4% at 120 min. The results indicated that the irradiation leads to the rupture of amorphous cellulose chains. The disintegration of amorphous regions may be explained by acoustic cavitation. As native BC in a liquid medium was exposed to ultrasonic irradiation, the acoustic waves induce alternating high and low pressure; this creates bubbles (i.e., cavities) and makes them oscillate. A bubble can grow while absorbing the ultrasonic energy at each cycle, until it becomes unstable and finally collapses violently, releasing the energy stored within it, subsequently producing shock waves in the medium [58]. A shear deformation during the collapse of the bubbles is considered to be responsible for the chemical effects which induce disintegration of the amorphous regions of cellulose.
Cavitation occurs over a very wide range of frequencies, from 10 Hz to 10 MHz. Above that frequency regime, the intrinsic viscosity of liquids prevents cavitation from occurring. According to Suslick and coworkers [59], most high intensity ultrasonic horns operate within the range of 20 to 40 kHz. Several factors can affect acoustic cavitation, such as reaction temperature, hydrostatic pressure, frequency, acoustic power, and the type of the solvent medium used. In our study, the amorphous regions degradation increases slightly after 30 min irradiation, and the CrI of cellulose for all sample experiments is still lower compared with that of commercial MCC (75.0%). In our experiments, optimum crystallinity is observed with 30 min irradiation; then, when reaction time is increased, degradation might occurred in both the amorphous and crystalline regions, consecutively reducing the product crystallinity. Concerning the solvent medium, when more volatile solvent is used such as ethanol in water, the mixture is expected to produce more cavitation bubbles, which can significantly promote the reduction rate of amorphous regions. We therefore compare our results (using water/ethanol 50% w/w) to reactions done in pure water. From XRD spectra, it was calculated that the increase in CrI was less pronounced when the reaction was run in sole water (increase from 60.7 to 67.0%) than in water/ethanol mixture (increase from 60.7 to 69.1%). This result is in accordance with DSC observations, as the endothermic peak of sample that was irradiated with ethanol (the red line in Figure 5B) was detected broader than the one with pure water (blue line in Figure 5B). Both samples have the same melting temperature, i.e., around T m = 344 • C. The endothermic peak for sample as detected in the DSC curves becomes larger when the crystallinity of the sample increases. According to Ciolacu et al., 2011 [57], the broadening of endothermic peaks detected in the DSC curves of celluloses is in a linear relationship with the percentage value of the amorphous material from their crystalline structure. absorbing the ultrasonic energy at each cycle, until it becomes unstable and finally collapses violently, releasing the energy stored within it, subsequently producing shock waves in the medium [58]. A shear deformation during the collapse of the bubbles is considered to be responsible for the chemical effects which induce disintegration of the amorphous regions of cellulose. Cavitation occurs over a very wide range of frequencies, from 10 Hz to 10 MHz. Above that frequency regime, the intrinsic viscosity of liquids prevents cavitation from occurring. According to Suslick and coworkers [59], most high intensity ultrasonic horns operate within the range of 20 to 40 kHz. Several factors can affect acoustic cavitation, such as reaction temperature, hydrostatic pressure, frequency, acoustic power, and the type of the solvent medium used. In our study, the amorphous regions degradation increases slightly after 30 minutes irradiation, and the CrI of cellulose for all sample experiments is still lower compared with that of commercial MCC (75.0%). In our experiments, optimum crystallinity is observed with 30 min irradiation; then, when reaction time is increased, degradation might occurred in both the amorphous and crystalline regions, consecutively reducing the product crystallinity. Concerning the solvent medium, when more volatile solvent is used such as ethanol in water, the mixture is expected to produce more cavitation bubbles, which can significantly promote the reduction rate of amorphous regions. We therefore compare our results (using water/ethanol 50% w/w) to reactions done in pure water. From XRD spectra, it was calculated that the increase in CrI was less pronounced when the reaction was run in sole water (increase from 60.7 to 67.0%) than in water/ethanol mixture (increase from 60.7 to 69.1%). This result is in accordance with DSC observations, as the endothermic peak of sample that was irradiated with ethanol (the red line in Figure 5B) was detected broader than the one with pure water (blue line in Figure 5B). Both samples have the same melting temperature, i.e., around Tm = 344 °C. The endothermic peak for sample as detected in the DSC curves becomes larger when the crystallinity of the sample increases. According to Ciolacu et al., 2011 [57], the broadening of endothermic peaks detected in the DSC curves of celluloses is in a linear relationship with the percentage value of the amorphous material from their crystalline structure.

Effect of MnCl2 Concentration on The Extraction of Crystalline Regions
The microwave treatment assisted by HCl-MnCl2 catalyzed hydrolysis was evaluated to hydrolyze cellulose. In this part, the use of a microwave reactor was performed to get a higher conversion and a shorter reaction time for catalyzed hydrolysis of depolymerized cellulose. In comparison, a conventional heating microwave is a high-frequency radiation that possesses both

Effect of MnCl 2 Concentration on The Extraction of Crystalline Regions
The microwave treatment assisted by HCl-MnCl 2 catalyzed hydrolysis was evaluated to hydrolyze cellulose. In this part, the use of a microwave reactor was performed to get a higher conversion and a shorter reaction time for catalyzed hydrolysis of depolymerized cellulose. In comparison, a conventional heating microwave is a high-frequency radiation that possesses both electrical and magnetic properties [60]. Regarding the addition of catalysts, MnCl 2 , was utilized to improve the extraction rate of crystalline regions during the hydrolysis. It was already shown that metal chlorides, due to their Lewis acid property, exhibit higher catalytic activity than inorganic acids [61]. The concentration of HCl used in this work was significantly reduced to 0.1 M, instead of 6 M, as used in many reported works. Table 4 shows the effect of the metal chlorides catalyzed hydrolysis reaction on the crystallinity index. In the absence of a catalyst (0%) in the 0.1M HCl medium, thermal hydrolysis could not occur effectively, and the CrI obtained was only 0.5% higher than for the starting material (DP-BC) from 69.1% to 69.6% for 30 min of reaction. Conversely, with the addition of MnCl 2 , the CrI was increased to 71.3%, 72.7%, and 79.4% for 1%, 2.5%, and 5% (w/w), respectively. A similar result was obtained for the use of FeCl 3 ·6H 2 O, 5% (w/w) as catalyst with a CrI increased from 69.1% to 77.8%. It was also found that the CrI of all experiments showed a lower value compared with that of the commercial NCC (85.0%). Nevertheless, it appears that the presence of a catalyst plays an important role in the extraction of crystalline regions. In our hypothesis, during partial depolymerization, the ultrasonic treatment leads to the distortion of the amorphous parts and eases the accessibility of chemical reagent to loosen them. Thus, the protons could more easily penetrate into the disordered regions during catalyzed hydrolysis, and as a result, greatly promote the hydrolytic cleavage of glycosidic bonds even in the diluted HCl medium. For this step, the hydrolysis reaction at 0.1M of HCl and 5% w/w of both metal chlorides (MnCl 2 and FeCl 3 ·6H 2 O) for 30 min reaction can enhance the removal of the amorphous regions, even though the CrI obtained is still less than 80%. The rapid degradation of amorphous regions during catalyzed hydrolysis can be explained by the Lewis acid character. According to Stein et al., 2010 [62], some metal chlorides such as FeCl 3 , AlCl 3 , CuCl 2 , and MnCl 2 could form hydrated complexes in aqueous solution and coordinate the glycosidic oxygen of cellulose. This helps to scissor the glycosidic linkages and to facilitate the hydrolysis process, while the chloride anions attack the hydroxyl atoms [61,63,64]. Introducing metal chloride salts into acid solution can further improve catalytic performance at which the intra-and inter-molecular hydrogen bonds can be broken, and the degradation of the amorphous regions can be accelerated by permeating the internal structure of irradiated cellulose (DP-BC) to acid. Moreover, under microwave and hydrothermal conditions, the easy diffusion of metal cations and chloride anions into the hydrogen bond network, as well as the strong ability of chloride anions to disrupt the hydrogen bond, can be achieved; thus, the hydrolysis rate is greatly enhanced.
Considering the thermal behavior analysis, the summary of DSC results is presented in Table 5 as follows: It was found that compared with the DP-BC, the degradation temperature of the BC-NC decreased by approximately 62.7 • C. Similar results were obtained for commercial MCC and NCC sample references. The nano-sized NCC exhibited lower degradation temperature than the micro-sized MCC, i.e., by 53.7 • C. The reason is that the thermal stability of nanocrystals is related to several factors including their dimension, crystallinity, and composition, which in turn depend on extraction conditions [65,66]. So, the NCC with the highest crystallinity would exhibit the highest thermal stability, but smaller dimensions should also cause a decrease of the degradation temperature. The FT-IR analysis demonstrated that the chemical structures of BC-NC remained unchanged after MnCl 2 -catalyzed hydrolysis process.

Conclusions
In this study, BC-NC was conveniently synthesized by sequential ultrasonic irradiation and microwave treatment. A simple and an eco-friendly approach was developed to control the degradation of bacterial cellulose at very low hydrochloric acid concentration. The results demonstrated that ultrasonic irradiation of BC in water/ethanol mixture led to cellulose depolymerization. The CrI increased from 60.7% to 69.1% during the first 30 min of irradiation. Yet, amorphous region distortion remained constant or increased slightly after 2 h, i.e., by 71.4%. It was found that microwave treatment using MnCl 2 as Lewis acid exhibited excellent catalytic activity and promoted the hydrolysis in diluted HCl. The reaction rate and the selectivity of BC-NC formation are shown to be optimal at 0.1 M of HCl with 5% w/v of MnCl 2 for 30 min. These conditions can enhance the removal of amorphous regions, yielding BC-NC that possesses an initiating decomposition temperature of 318.6 • C, and led to improve the CrI of up to 79.4%. The BC-NC is typically a mixture of small needles of (164.51 ± 7.56) nm in length and (25.05 ± 2.80) nm diameter that can form fibrillated structures. This procedure yields nanocrystalline bacterial cellulose having similar features to commercially-available nanocrystalline cellulose.