Abstract

Nanocrystalline cellulose (NCCs) was synthesized from pruning waste of Zizyphus spina christi using H₂SO₄ (64 %, wt/wt) under suitable hydrolysis conditions. The crystal growth of the NCCs from nano- into identical micrometric-scaled needles confirmed their ability to self-assembly. The aspect ratio of the NCCs was estimated using optical microscopy for needles, and by scanning electron microscopy (SEM) for powder, while their crystallinity index (CI), crystallite size (CS) and lattice spacing (LS) were estimated by X-ray diffraction (XRD). Thermogravimetric analysis (TGA) and differential thermal analysis (DTA) were also performed. The XRD-diffractogram of the NCCs was similar to that known for cellulose I. The CI of the NCCs was much higher (86.75%) than that for cellulose I. The CS of the NCCs was 2.78 nm that is smaller than that for cellulose I. The distance between the strata within the NCCs (LS) was found to be 0.214 nm. The TGA indicated a gradual increase in the mass loss upon heating the NCCs from 25°C up to 500°C in a flowing N₂-atmosphere. The DTA showed presence of an endothermic peak (due to H₂O-evaporation) and one exothermic peak (due to depolymerization and decomposition of the NCCs. Based on the results, the Zizyphus wood is suitable precursor for the NCCs production.

1. Introduction

Cellulose is the most abundant biopolymer on the Earth. By dissolving amorphous regions of microfibrils, the crystalline components were released. Colloidal NCCs rods having high aspect ratio have been attracting wide acceptance in global markets due to their unique mechanical properties, chirality, sustainability, and availability ¹.

NCCs production is a depolymerization process of cellulosic fibers into a value-added nanomaterial, or producing useful chemicals and fuels ². In relation to the hydrolyzing agent, the NCCs were synthesized by using H₂SO₄ ³, HCl ⁴, HBr ⁵. In addition, phosphotungstic acid (H₃PW₁₂O₄₀) was used by Lu et al. ⁶ as a catalyst for hydrolysis of cellulose under mechano-chemical activation. Furthermore, Chen et al. ² used Cr(NO₃)₃ as a hydrolyzing agent, while TEMPO-mediated oxidation of NCCs in alkaline conditions was applied by Sadeghifar et al. ⁵.

Beck-Candanedo et al. ⁷ indicated that the hydrolysis duration is one of the most important parameters affecting the hydrolysis products of wood pulp. The effect of the reaction conditions on NCCs-surficial charge and sulfur content was not significant and was controlled by factors other than hydrolysis conditions. However, chiral nematic pitch decreases when increasing the cellulose concentration and decreasing the NCCs length. Accordingly, aspect ratio of the NCCs also affects their chiral nematic property.

Araki et al. ⁴ found that H₂SO₄ provided more stable aqueous suspensions of NCCs than did HCl. The NCCs produced by HCl had minimum surficial charge, while those synthesized by H₂SO₄ were found to have a negatively charged surface ⁸. This is due to the esterification process occurred for their surficial hydroxyl groups to give charged sulfate groups ⁷.

In addition, Bondeson et al. ⁹ emphasized the importance of duration and temperature of hydrolysis together with the H₂SO₄ concentration as important singular factors in the process of preparation of negatively charged isolated cellulose whiskers in water. Cellulose whiskers with a length ranging between 200 and 400 nm were obtained by using H₂SO₄ (63.5 wt/wt %) for approximately two h and with a yield of 30%.

An efficient green high-yield approach was applied by Lu et al., ⁶ to produce high yield of NCCs (up to 88.4%) using H₃PW₁₂O₄₀-catalyst through a mechano-chemical hydrolysis.

Crystallographic scientists have controlled several properties of the NCCs such as composition, size, shape, structure and surficial properties ¹⁰. Geometrical characteristics such as size, dimensions and shape of NCCs depend on the nature of the cellulose source as well as the hydrolysis conditions such as duration, temperature, ultrasonic treatment, and purity of materials ^{7, 11, 12}. Above a critical concentration point, the rod-like shape of the charged NCCs leads to the formation of an anisotropic liquid crystalline phase ^{7, 13}. Nevertheless, typical dimensions of whiskers range from 5 to 10 nm in diameter and from 100 to 500 nm in length.

In a stable cellulose nanofibrils (CNF) suspension, the distances among fibrils are not small enough to form molecular contact. Upon drying, forces resulting from the removal of water and high temperatures may drive the molecular contact of the CNFs and subsequently cause agglomeration. There are four methods can be performed to dry the NCCs, namely freeze drying, spray-drying, oven drying and supercritical drying ¹⁴.

There are several methods for calculating CI from XRD data, including the peak height method (PHM), Jayme-Knolle method (JKM), and modified Ruland-Vonk method (MRVM). Furthermore, Smoothing and differentiation are necessary in the XRD analysis to separate the true effects of the X-ray on a material’s molecules ^{15, 16}. Materials with high crystalline cellulose content (with high CI) give a sharp peak with a high intensity, whereas those with large amounts of amorphous components such as lignin, hemicelluloses, pectin, and amorphous cellulose (with low CI) give a broad peak with a low intensity ^{17, 18}. The crystallographic characteristics can be investigated by various techniques and computation methods using XRD, ¹³C-nuclear magnetic resonance (¹³C-NMR), Fourier transform infrared (FTIR), and Raman spectroscopies ^{19, 20, 21}.

The CI of cellulose is differed according to the cellulosic precursor, smoothing algorithms used for handling the XRD data, calculation methods and measurement technique used ^{15, 19, 20, 22}. It was found by Hindi ²³ that the XRD-diffractograms of two seed flosses (Ceiba pentandra and Calotropis procera) had a broad peak at 2θ of 18° for amorphous components and a sharp peak at approximately 22° related to crystalline components. Furthermore, the CI of seed floss of Calotropis procera was 65.5%, which approaches that for wood pine (70%) indicated by Borysiak and Doczekalska ²⁴ and lies within the CI range (41.5% to 95.5%) shown by Park et al. [2010], (56% to 78%) by Terinte et al. ²⁰, 79.6% obtained by Lu et al., ⁶ and 86.5% found by Chen et al. ² for different cellulosic precursors using different measuring techniques. The higher crystallinity investigated by the wide angle-XRD for Calotropis procera seed floss than that for Ceiba pentandra reflects lower amorphicity for the former resource in its fiber wall nanostructure, which offers an additional mechanism for the latter to sorb more fluids ²⁵. The CI of Avicel cellulose differed significantly, from 39% to 67% according to the measurement technique used ²². The ranges of the reported CI were 56% to 78%, 65% to 83%, and 37% to 93% for cotton linters, microcrystalline cellulose, and Avicel PH – 101, respectively ²⁰.

Lu et al., ⁶ produced NCCs with higher thermal stability by combining mechanochemical activation and phosphotungstic acid hydrolysis. It was emphasized by Chen et al., ² that better thermal stability of NCCs synthesized by Cr(NO₃)₃ (344°C) compared to those produced by H₂SO₄ (273°C) will force the former material to be regarded as a high potential filler for industrial nanocomposite applications.

It was reported by Dufresne ¹¹ that neutralization of cellulosic nanoparticles, synthesized by sulphuric acid, by using NaOH can enhance their thermal stability.

2. Materials and Methods

The study was performed during 2015-2016 in the Central Laboratory at the Agricultural Research Station (ARS), Hada Al-Sham, King Abdul-Aziz University, about 120 km northeastern of Jeddah.

Woody branches of Zizyphus spina var christi (L.) Desf. was chosen for cellulosic fibers maceration and subsequent NCCs-production. Four healthy Zizyphus shrubs of 12 years old-stand were chosen randomly and three branches of each tree were selected. The branches were pruned in April by cutting at 5 cm above their branching level and specified for the maceration process. The mean diameter outside bark varied from 5 to12 cm.

From each of the selected branches, one disc (about 20 cm along its grain) was removed at about 10 cm above the branching level. After excluding the pith and bark, the remainder wood of the disc was cross-cut into cubes (2 cm long each), Then, wood cubes were converted into thin chips (2 cm×2 cm×1 mm thick). The wood samples were extracted through a tertiary stage by a mixture (1:2) of ethyl alcohol (95 %)-benzene, ethyl alcohol (95 %), and finally by hot water for three hours in a Soxhlet apparatus according to ASTM ²⁶ and Hindi et al. ²⁷.

Five grams of thin chips from each of the three branches of each tree were digested separately using the Franklin method using a mixture of hydrogen peroxide (35%) and glacial acetic acid in a ratio of 1: 1, and kept at 60°C until the clear whiteness aspect. The oven-dried macerated fibers were used as a precursor for the NCCs synthesis.

The NCCs were synthesized from the macerated cellulosic fibers by H₂SO₄, 64 % w/w ³ at 70°C with continuous stirring for an hour. The reaction was quenched by diluting the solution up to 20 fold by deionized water. The suspension was centrifuged at 1500 rpm to remove unhydrolyzed fibers and then at 14000 rpm for 20 min to obtain the NCCs. The precipitate was collected, re-centrifuged, and dialyzed against deionized water until neutralization. No sonication exposure was done upon the NCCs synthesis.

The studied traits for the oven-dried NCCs particles were length and width that were used to calculate their aspect ratio. Moreover, the crystallinity index (CI), crystallite size (CS) and lattice spacing (LS) by XRD, mass loss by thermogravimetric analysis (TGA), and the difference in the sample’s enthalpy (as exothermic or endothermic reactions) via differential thermal analysis (DTA) were applied to characterize the NCCs properties.

An optical speculation unit consisted of a light microscope (CE- MC200A) in a magnification power of 10X with suitable vision system (OPTIKA PRO 5 Digital Camera- 4083.12, OPTIKA, Italy) with a Vision PRO 4 software was used to pick up, processing images, record different measurements of the NCCs-needles and monitoring their crystal growth ^{3, 28}.

One drop of recently prepared nanometric NCCs supernant, immediately at the end of the hydrolysis duration, was mounted and spread onto a glass slide without staining. One slide was specified to represent each of the four replicates of the wood species. The aspect ratio (length/width) of the NCCs-needles was calculated. About twenty needles were observed from each slide.

The oven-dried NCCs powder was redissolved in deionized water and sonicated using a digital ultrasonic Cleaner HB-4818 T. One drop of the previously filtered NCCs supernant was mounted and spread onto a double side carbon tape on Al-stub and dried. Before testing, the samples were sputtered with a 15 nm thick gold layer (JEOL JFC- 1600 Auto Fine Coater) in a vacuum chamber ²⁹. The samples were examined with a SEM Quanta FEG 450, FEI, Amsterdam, Netherland at an accelerating voltage differing between 5-20 kV. The length and width of the oven-dried NCCs particles were measured and the aspect ratio of them were calculated.

The wide-angle XRD spectra of the oven-dried NCCs particles were obtained using an XRD 7000 Shimadzu diffractometer, Japan ¹⁵. For the XRD system, CuKα radiation was consisted of Kα1 (0.15406 nm) and Kα2 (0.15444 nm) and was induced at 30 kV and 30 mA. The CuKα radiation was extracted from the data using a single-channel analyzer. Each divergence and scatter slits was 1°, and the receiving slit was 0.15 mm at the same radius. Oven-dried NCCs samples (about 0.5 g) were mounted onto quartz using a dry amorphous glue. The two theta angle range of 10°-30° was used to study the crystallographic properties of the NCCs. All runs were operated in the reflection mode with scanning velocity of 4°/min in 0.05°-steps.

When the cellulosic sample was X-rayed, a stream of crude data representing x- and y-coordinates (2θ and I, respectively) of all points representing the XRD diffractometer was obtained (Figure 4). The XRD data were processed using curve table-software for smoothed data fitted by the Savitzky-Golay algorithm.

After the individual crystalline peaks were smoothed, the CI was calculated as a percentage of the crystalline diffractogram area based on the original diffractogram area using the following equation ¹⁵:

where D_cr1 is the area under the 1^st crystalline peak representing the 110 and 11¯0 planes together, D_cr2 is the area under the 2^nd crystalline peak related to the 200 crystallographic plane, and the D_t is the area under the total diffractogram.

The CS was calculated by the Scherrer equation ^{30, 31} as follows:

where CS is the average diameter of the crystallite, k is the form factor of the crystallite (0.94), λ is known to be wavelength of X-ray accident to the crystallite (0.1542 nm), β is the full width at half maximum (FWHM) of the crystalline peak represented to the crystallographic plane 200, and θ is the Bragg’s angle related to the 002-plane. A factor of 57.3 is multiplied by kλ to convert β from degrees to radians in the above mentioned equation.

The d value was calculated using the Bragg’s equation mentioned below:

where n is an ordinal number taking a value of “1” for diffractograms having the strongest intensity [66], λ is the wavelength of X-rays incident on the crystal (0.1542 nm), and θ is the Bragg’s angle corresponding to the 200-plane ¹⁵.

Analyzing the thermal properties of the NCCs is important to monitor their processing temperature range and utilization. In the present investigation, the thermal analysis was restricted to two characterization for the resulted NCCs, namely thermogravimetric analysis (TGA) and differential thermal analysis (DTA) using a Seiko & star 6300 analyzer, Central Laboratory, Faculty of Science, Alexandria University, Egypt. Heating was scanned from 30°C up to 500 °C with an increasing rate of 20 °C/min in a flowing nitrogen atmosphere ^{3, 32}.

The mass loss of the NCC was calculated from the TGA curve presented in Figure 5 using the following equation:

Mass loss=[(W₂-W₁)/W₁]× 100, where

W₁= Initial NCCs weight for such a temperature region.

W₂= Final NCCs weight for the same temperature region.

Completely randomized block design with four replications was performed in this study using the analysis of variance method. Furthermore, the differences between means were examined using the least significant difference test (LSD) at P≤0.05 according to Steel and Torrie ³³.

The target of this work is to evaluate the quality properties of the nanocrystalline cellulose (NCCs) synthesized from cellulosic fiber macerated from Zizyphus spina christi wood.

3. Results and discussion

The OM findings indicated some dimensional insights corresponding to the needles architectures constructed upon crystal growth of the nanometric-scaled NCCs-acidic solution. This growth from nano- into micrometric scale showed the ability of the NCCs for self-organization or so-called self-assembly. It was noticed that after about 30 minutes after spreading a drop of the hydrolysis supernant, immediately after completing the hydrolysis process, onto a glass substrate, the NCCs spherulites were agglomerated into bigger aggregates belongs to colloidal state which in turn were aligned straightly to form needle-shaped architectures. The driving force for this crystal growth process can be restricted to the following effects: a) electrostatic forces on the NCCs surfaces due to the grafted functional groups (protons, sulfate and hydroxyl), b) the H₂SO₄-concentration gradient, and c) the difference between surface tensions of solution, air and glass. The net force generated at this interface is considered as the driving force of the NCCs-crystal growth process. Accordingly, the needles were imagined to be formed by electrostatically end to end attraction and subsequent self-welding (Figure 2).

The SEM micrograph of the NCCs particles is presented in Figure 3. It is clear that the NCCs are finally aggregated, after oven-drying, in a short rod-shaped with a vast range of particle size distribution. The final rod lengths varied from 1.875 μm to 26.25 μm, while their widths ranged from 0.75 μm to 5.25 μm.

The aspect ratio of the NCCs-Needles was ranged from 50 to 100 based on the optical microscopy speculation (Figure 2). For the aspect ratio of the oven-dried NCCs particles based on the SEM study, it was calculated to be 2.5-5. The aspect ratio of the NCCs-needles was about 20 times higher than that for the NCCs-powder. The high aspect ratio of the NCCs-needles add a valuable feature for them, but unfortunately, this architecture is unstable and are fractured into short rods with the evaporation of the acidic media or under neutralization and filtration process.

For the aspect ratio of the oven-dried NCCs particles, it was smaller than many literatures such as those found by Tonoli et al. ³⁴ for micro/nanofibrils (10-50), and Sacui, et al. ³⁵ for NCCs (23±12-148±147). Since colloidal NCCs rods having high aspect ratio have been getting wide acceptance in the global markets due to their exquisite mechanical properties, chirality, sustainability, and availability ¹, the oven-dried NCCs must be converted into nanocolloidal form to be included in a composite material to enhance their aspect ratio.

The NCCs samples exhibited a principle sharp peak around 2θ = 21.11^o representing the 200 reflection related to hemicelluloses and alpha-cellulose (Figure 4).

In addition, the NCCs sample showed two broad peaks at 2θ = 18.32^o and 19.28^o representing 110 and 11¯0 reflections. Accordingly, the similarity between the resultant NCCs and cellulose-I was clear, especially when regarding the crystallographic planes, namely 110, 11¯0 and 200 ^{36, 37, 38} as indicated in Figure 4.

The CI of the NCCs was found to be high (86.75%) indicating that the CI of this material was increased after removing the amorphous component via acid-hydrolysis. The obtained CI was higher than those obtained for cellulose and NCCs (70.62 and 76.01%, respectively) that found by Wulandari et al. ³⁹, seed floss of Calotropis procera (65.5%) ²³, wood pine (70%) that indicated by Borysiak and Doczekalska ²⁴, 79.6% that obtained by Lu et al. ⁶, and lies within the CI ranges (41.5% to 95.5%) that shown by Park et al. ¹⁹, and (56% to 78%) determined by Terinte et al. ²⁰, and approaches to the 86.5% value found by Chen et al. ² for different cellulosic precursors using different measuring techniques.

The crystallite size (CS) is the crystallite thickness estimated by the Scherrer formula for the crystallites having a size less than 100 nm ³⁰. The average CS of the NCCs was measured to be 2.78 nm that is similar to that found by Hindi ³, and smaller than that for cellulose I (about 5 nm in width). The CS finding is in agreement with the data range found by Hindi ¹⁵. It is worth mentioning that, XRD resolution is not adequate for small crystallites to obtain accurate imagination concerning their lattice structure ⁴⁰.

The LS of the NCCs is a measure of the distance between their successive cellulosic strata within a crystallite ⁴⁰. Using the Bragg’s equation, the distance between the strata was found to be 0.214 nm. Since larger crystal size leads to larger LS between its crystalline strata ⁴¹, the lower LS value can be attributed to the small size of the NCCs crystallite estimated in the present study (2.78 nm). The LS result is slightly smaller than that found by Hindi ¹⁵.

The TGA thermogram (Figure 5) clearly indicates a gradual increase in losing weight of the NCCs upon heating them from 25°C up to 500°C in an inert atmosphere of a flowing N₂-gas. The entire heat range was divided into five individual regimes, namely 25-100°C, 100°C-200°C, 200°C-300°C, 300°C-400°C, and 400°C-500°C to study the mass loss of the NCCs occurred at each regime. The primary mass loss for the NCCs was 5.26 % as temperature increased from 25°C to 100°C due to elimination of free water ³. In addition, the NCCs have lost about 22.3 % of their mass when the temperature was raised from 100 up to 200°C due to evaporation of both hygroscopic and constitutional water. Between 200°C-300°C, the NCCs mass loss was lower than that recorded for the 2^nd thermal degradation regime (12.24 %). Afterward, the NNCs was continued to lose part of their mass as 25% and 16.7 % for the 4^th (300°C-400°C) and 5^th (400°C-500°C) regimes, respectively. It can be seen from Figures 5 and 6 that the mass loss was shifted at the 2^nd and the 4^th regimes comparing to the other thermal degradation regimes. At lower temperatures, the thermal degradation of the sulphated NCCs can be attributed to presence of huge free ends in chains of the highly sulfated regions (HSR). It worth mentioning that the HSR have less crystalline nature that makes them more susceptible to thermal elimination ⁴² as volatile byproducts and may lead to drastic reduction in the NCC-molecular weight. In addition, elimination of the sulphate groups in sulfated anhydro-glucose units require less energy ⁴³, therefore, sulfuric acid molecules were arisen at lower temperatures upon the thermal degradation process.

On the other hand, the low thermal stability of the sulphated NCCs at higher temperatures can be related to the degradation of unsulphated components of the NCCs material ⁴⁴.

DTA is a measure tool to differentiate two hot materials, the 1^st one is the NCCs and the 2^nd one is an inert reference material. Both materials are found at the same place and conditions. As shown in Figure 7, there is an endothermic peak (under the baseline) as well as one exothermic peak (above this line). The endothermic peak starts from 39°C up to 277°C with a maximum value at 150°C, while the exothermic peak begins from 277°C until 460°C with a maximum shift at 356°C (Figure 7 and Table 1).

The endotherm of the DTA results can be attributed to evaporation of the three forms of moisture content in the NCCs, namely free, hygroscopic and constitutional moisture along with fusion or melting process of the NCCs. Furthermore, losing moisture is companied to this endotherm at higher temperatures ³⁶. The evaporation process depends essentially on the difference in holding capacity of moisture by the sorptive forces ³⁶ arisen between water molecules and the NCCs-crystallites. For the sulphated NCCs, the hydrolyzing agent used in this study (H₂SO₄) acted as a dehydrating agent and helped for grafting the sulphate groups which are well-known by their hydrophobicity that lowering the NCCs-affinity for moisture sorption. It worth to mention that after hydrolysis of the macerated cellulosic fibers, all amorphous regions were dissolved and only nano-scale cellulose crystals would possess sulphate groups that responsible for the higher onset temperature of crystal melting with wider endotherm width.

The exotherm occurred between 277°C and 500°C can be attributed to depolymerization of the sulphated NCCs, decomposition of glycosyl-units and then formation of a carbonaceous residues (36). This depolymerization can be illustrated by four reasons: a) due to nano-size and greater number of the free ends of the chains of NCCs which decompose at a lower temperature, b) H₂SO₄ (a dehydrating agent) which facilitate the depolymerization of cellulose by removing some hydroxyl groups ⁴⁵, c) the presence of H⁺ ion in the weak acidic atmosphere (pH 5.5-6) that increased the char residues because of removal of oxygen in the form of H₂O which prevent weight losses, and (d) the highly crystalline nature of CNCs increased the carbon residues ^{38, 45}.

Furthermore, the absolute value of the heat change values for the endotherms was 639 μVs/mg that is extremely higher than that released due to the exotherm (37 μVs/mg). Since the material absorbing higher energy are more thermally stable than those absorbing lower energy or releasing more energy upon exothermic reactions ³², the NCCs have a good thermal stability behavior.

4. Conclusion

• Crystal growth of the nanocrystalline cellulose from nano-into more organized-micrometric-scaled architectures confirmed their ability to self-assembly.

• The nanocrystalline cellulose spherulites were agglomerated into bigger aggregates upon crystal growth on a glass substrate in an acidic media.

• The driving forces for the crystal growth process are electrostatic forces on the crystallite surfaces, the H₂SO₄-concentration gradient, and the difference between surface tensions of solution, air and glass.

• The nanocrystalline cellulose diffractogram was similar to that known for cellulose I.

• The crystallinity index of the nanocrystalline cellulose was higher than that for cellulose I confirming elimination of amorphous components due to acid-hydrolysis.

• The thermogravimetric analysis indicated a gradual increase in their mass loss of the synthesized NCCs upon heating them from 25°C up to 500°C in a flowing N₂-atmosphere.

• Differential thermal analysis showed the presence of an endothermic peak (due to evaporation) and one exothermic peak (due to depolymerization and decomposition of the nanocrystalline cellulose

• Based on the results, the Zizyphus wood is suitable precursor for nanocrystalline cellulose production.

Acknowledgments

This project was funded by the Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah under grant no. 115/155/1432. The author therefore, acknowledges with thanks DSR technical and financial support.

References