Insight into the extraction and characterization of cellulose nanocrystals from date pits

2.1 Materials

Qatari date fruits, Phoenix dactylifera L. were collected from local stores in Qatar. Sodium hydroxide (research lab fine chem industries, India), sodium hypochlorite (Alfa Aesar, the United States), n-hexane (Scharlab S.L., Spain), and sulfuric acid (Sigma-Aldrich, United States) were of analytical grade and used without any further purification. Distilled water was used in the whole process. Dialysis bags were provided by Carolina Biological Supply Company.

2.2 Preparation of starting material

The date pits (DP) samples were washed with distilled water to remove any dirt and impurities. The samples were left overnight to air dry. A microphyte disintegrator was utilized for the grinding of DP. After that, the samples were stored at room temperature for future use.

2.3 Extraction of cellulose and cellulose nanocrystals from date pits

The schematic illustration of the used extraction methods for cellulose nanocrystals (CNCs) from the DP is shown in Fig. 1.

Close

Fig. 1

Extraction methods of CNCs from the DP.

Export to PPT

2.4 Preparation of cellulose nanocrystals (CNCs)

According to our preliminary research, the obtained cellulose from methods 1 and 2 was subjected to 500 mL of 64% sulfuric acid (H₂SO₄) for 1 hr under 25 °C (Wahib et al., 2022). The reaction ended by adding cold distilled water that is 10 times the volume of the reaction. Then, neutral pH was achieved by repetitive washing of sulfuric acid with distilled water. After that, separation of CNCs has been carried out by using the centrifuge for 30 mins at 5000 rpm. Moreover, the excess sulfate ions were removed by using a dialysis membrane in which the CNCs were soaked in distilled water for 48 hrs, which then has been followed by ultrasonication for another 15 mins at 600 w. Finally, an oven was used to dry the obtained CNCs at 70 °C overnight. The CNCs obtained by method 1 will be denoted as CNCs₁ and by method 2 as CNCs₂.

2.5 Characterization of CNCs and intermediate products

2.6 Cost analysis

Cost analysis was done to compare the two methods that were used for CNCs extraction from DP. The analysis was carried out by breaking down the cost of the used chemicals, materials, and energy consumed as well as other extra expenses.

3.1 Chemical components

Table 2 shows the chemical compositions of N₁-DP, B₁-DP, N_2-DP, B_2-DP, and DP. The chemical compositions of the intermediate products of the two methods were found to be significantly different.

The results of the chemical compositions of the DP at different treatment stages of methods 1 and 2 are listed in Table 2. It is important to note that there are other chemical components such as protein, fat, and ashes found in DP. Nevertheless, these components do not affect the isolation of CNCs due to their low percentages as reported by Genedy et al. (2017). The DP has a chemical composition of 21.2, 28.1, and 19.9% for α-cellulose, hemicellulose, and lignin, respectively (Al-Ghouti et al., 2019). The non-cellulosic content like hemicellulose has the highest percentages in comparison to the cellulose content. Therefore, the objective was to chemically treat the DP to remove the non-cellulosic contents. Hence, when the DP was subjected to treatment by NaOH, the hemicellulose and lignin contents of the DP decreased from 28.1% and 19.9% to 7.95% and 6.7%, respectively for both methods (N_1-DP & N_2-DP). On the other hand, the concentration of cellulose increased for both methods (N_1-DP & N₂-DP) to 63.1% and 51.5%, respectively. Comparing these values to the original α-cellulose of the DP, which is 21.2%, the values bare significantly high. Hence, the delignification process performed by the NaOH treatment was successful. Furthermore, it can be concluded that method 1 yields better results than method 2, in terms of chemical composition after the delignification process.

After the bleaching treatment, the hemicellulose and lignin contents decreased even more to 2.6% and 1.5%, respectively for both methods (B_1-DP & B_2-DP). Abu-Thabit et al. (2020) extracted and characterized microcrystalline cellulose from Phoenix dectylifera L. date seeds and found that bleached date seed has 3.92% hemicellulose and 0.86% lignin. Moreover, it can be justified that the bleaching treatment is the second treatment for the DP, therefore less hemicellulose and lignin contents are yielded. Whereas it is evident that the cellulose content for both methods (B_1-DP & B_2-DP) increased significantly to 84% and 72.5%, respectively. This can be due to the amorphous phase removal which was indicated by the presence of alkali-soluble lignin components as well as the water-soluble hemicellulose, which in turn led to the presence of higher cellulose content (Tarchoun et al., 2019). It is clear that the first method that relies on the mechanical stirrer yields better results for purified cellulose.

3.2 Physicochemical properties of CNCs₁ and CNCs₂

The physicochemical properties of CNCs₁ and CNCs₂ are listed in Table 3. Measurement of the true density of cellulosic materials is important since it provides the required physical quantity value needed for future practical uses. The density of CNCs₁ (1.7) and CNCs₂ (1.9) are considered to be very close in value to each other. Vincent et al. (2021) used postharvest waste (stalks) to extract CNCs and found that it has a density of 1.6. In addition, comparing these results to a previous study, the true density of nanocellulose was found to be almost 1.60 regardless of the cellulose's biological origin (Daicho et al., 2019). This means that most nano-scale products have true densities that are less than 2.0 irrespective of the chemical composition and other physiochemical properties the fiber possesses. According to Crouter & Briens (2014), the moisture content is considered a significant parameter when dealing with CNCs at pilot scales such as manufacturing stages. In the current results, the measured moisture content of CNCs₁ is 5.6% and CNCs₂ 6.2%. Evans et al. (2019) extracted CNCs from sugarcane baggase and found that it has a moisture content of 3%. It can be concluded that the moisture content of method 2 is slightly higher due to the extraction step performed with hexane and water, nevertheless, there are no significant differences between both methods. This might be due to the fact that extraction in method 2 using hexane and water was used to dissolve and remove impurities that alter the properties of CNCs surfaces, these impurities include hemicelluloses, lignin, and different mono and disaccharides (Banerjee et al., 2020). Therefore, the slight moisture content may be attributed to the absorption of moisture in the spaces left vacant from the removal of hemicellulose and lignin (Sofla et al., 2016). Lastly, the pH value for the CNCs₁ suspension (6.4) and that of CNCs₂ (6.2) are close to neutral, which fulfills our aim at obtaining pure cellulose nanocrystals after modifying by sulfuric acid. Similarly, Pandi et al. (2021) synthesized cellulose nanocrystals from cotton and found that the pH of the obtained CNCs suspension was between 6 and 7.

Furthermore, considering the yield of both CNCs samples as per equation (3), CNCs₁ has a slightly higher yield (12%) than CNCs₂ (10%). Nonetheless, both methodologies presented a yield (≥10%) which is considered a promising technique for CNCs production on an industrial scale (Seta et al., 2020).

3.3 Scanning electron microscopy (SEM) analysis

The DP, N_1-DP, B₁-DP, N_2-DP, and B₂-DP were observed using a scanning electron microscope as shown in Fig. 2. The SEM images of the DP demonstrate that the sample has agglomerated and smooth surface morphology with few rough areas, which is due to the presence of hemicellulose and cellulose content that compacts and hides the cellulose fibers below them. The high magnification of the SEM image of the DP surface reveals an extremely smooth surface, which can be due to the reflection of the lignin content that coats the cell wall of the DP. Comparing this to the intermediate products of both methods (N_1-DP, N_2-DP, B_1-DP, & B_2-DP), it is evident that all the samples exhibit a rough surface with irregular morphologies because of the strong chemical treatments during the delignification and bleaching process. However, after the delignification process by NaOH, it can be inferred that the surface of both N_1-DP & N_2-DP had an irregular structure with a porous exterior. In other words, the presence of the hollow morphological surface proves the removal of the filled parts of the surface as a result of the removal of the amorphous components. Similarly, Zou et al. (2020) studied the morphology of transparent cellulose nanofibrils composites and found that the removal of cementing amorphous components led to the presence of hollow surface morphology. This can be attributed to the degradation process of the hemicellulose and the lignin content that occurred by the alkali treatment (Fu et al. 2018). Further, B_1-DP and B_2-DP exhibited porous structures, as can be shown in Fig. 2e and f. The porous morphology of the bleached samples is clear and that is due to the successful removal of hemicellulose and lignin contents. Additionally, this indicates that the removal of the residual lignin by the NaClO solution was successful. All the obtained results are also consistent with the values reported in chemical composition in Table 2. Similar results were obtained by Zheng et al. (2019), who extracted CNCs from walnut shells and found that SEM results indicate the degradation of lignin and hemicellulose due to the alkali treatment. Furthermore, Vardhini et al. (2018) investigated the effect of alkali treatment of different concentrations on the morphology of banana fiber, and SEM results indicate the removal of hemicellulose and lignin.

Close

Fig. 2

Schematic illustration of SEM results of the surface morphology of A) DP at 5000x, B) DP at 2500x, C) N_1-DP, D) N_2-DP, E) B_1-DP, F) B_2-DP.

Export to PPT

3.4 Transmission electron microscopy (TEM) analysis

The transmission electron microscope was used to study and observe the morphology of the cellulose nanocrystals under the two different methods. The arrangement of the cellulose nanocrystals extracted after acid hydrolysis of bleached DP is shown in Fig. 3 of both CNCs₁ and CNCs₂. The TEM was the best to study the cellulose at the nanoscale. Based on literature and previous studies, the usage of sulfuric acid at optimum concentration, temperature, and time yields ideal nanocrystals (Khan et al., 2020; Guo et al., 2020; Zheng et al., 2019). The TEM results for the CNCs₁ morphology after modification through sulfuric acid hydrolysis show cellulose microfibrils with packed structure because of the presence of intermolecular and intramolecular hydrogen bonding (Chen et al., 2011). It could also be due to the sonication treatment that caused the fibers to spread unevenly, but not well enough, therefore at low magnification, multiple aggregations of nanofibers are present. It is important to note that when preparing nanosize particles sufficient sonication for the sample is required to fully break the fibers into smaller microfibers (Abu-Thabit et al., 2020). Therefore, the CNCs₁ sample needs to be further sonicated to avoid the aggregation of the nanofibrils. Furthermore, the appearance of the aggregated nanofibrils could be due to the high specific area between the nanofibers themselves. In addition, it is evident that the fibers have rod-like morphologies, which is one of the typical cellulose nanocrystals shape (Kontturi et al., 2018). The same findings of nanocellulose TEM images obtained through sulfuric acid hydrolysis are reported by Moreno et al., 2018. Nevertheless, at high magnification, the morphology of the nanofibers shows that the aggregates are assembled in parallel to each other and are aligned neatly alongside one another. This could be explained by the fact that the hydrogen bonds between the hydroxyl groups are found on the surface of the nanofibrils, which forces the fibers to be aligned in that way (Chen et al., 2011). Therefore, at high magnification, the appearance of the fibrils is of an organized structure. Furthermore, the ordered arrangement of the fibrils represents the crystalline region of CNCs₁ with few amorphous regions. This is due to the removal by sulfuric acid chemical treatment. It is worth mentioning that 64% concentration of sulfuric acid has been reported to be the best in the formation of nanocellulose and that using lower acid concentration can yield lower nanocellulose (Börjesson and Westman, 2015). As for the second methodology, it shows that the CNCs₂ also possesses a packed structure due to the strong hydrogen bonding. Moreover, the fibrils, in this case, are more dispersed and less clumped in comparison to CNCs₁, although it is similarly evident that the fibers are also rod-shaped cellulose particles. Although, as the magnification increased, the morphology of the nanofibers is shown to be assembled next to each other, where they are similar to the TEM image of CNCs₁. Both methods for CNCs₁ and CNCs₂ have similar morphologies, meaning that the extraction processes using both methods were applicable and successful. In the end, the size and shape of cellulose nanocrystals impact the properties it has in an aqueous media, hence these characteristics determine the application of CNCs.

Close

Fig. 3

TEM image of A) CNCs₁, B) CNCs₂.

Export to PPT

3.5 Fourier transform infrared (FTIR) analysis

The importance of the FTIR spectrum relies on the fact that it demonstrates significant information regarding the chemical structure of the biopolymers obtained. The presence of multiple functionalities such as hydroxyl, methoxy, and cellulosic components were verified from the characteristic peaks detected in the FTIR spectrum of both CNCs₁ and CNCs₂ as shown in Fig. 4. This form of analysis can give us an idea of whether the extracted cellulose nanocrystals are similar in their chemical structure or whether the different methodology of extraction can reveal different chemical structures of CNCs.

Close

Fig. 4

FTIR spectra of CNCs₁ and CNCs₂.

Export to PPT

Since cellulose, hemicelluloses, and lignin are the main constitutes of DP, therefore, the main functional groups present in these three materials are alkanes, esters, aromatics, ketones, and alcohols with different oxygen-containing functional groups (Chandra et al., 2016). As a result, the elimination of some peaks in the FTIR spectra will indicate the complete removal of the component's posts chemical treatment of the DP fiber. Two main absorption regions appeared in both curves, where one is at low wavelengths from 500 cm⁻¹ to 1750 cm⁻¹ and the other at high wavelengths from 2800 cm⁻¹ to 3500 cm⁻¹ (Zheng et al., 2019). Moreover, O-H stretching vibration in both samples that represent the hydrogen-bonded hydroxyl groups in the cellulose molecules is found near the region 3400 cm⁻¹ as a wide broad peak. The peak intensity is reduced due to the successive treatments made by alkali and acid treatment on the DPs to obtain cellulose nanocrystals. Moreover, it can be due to the degradation of the intramolecular and intermolecular hydrogen bonds formed by most of the hydroxyl groups (Abu-Thabit et al., 2020; George & Sabapathi, 2015; Adel et al., 2011). This can also result in lower moisture absorption, hence a broad wide peak is formed (Chandra et al., 2016; Abraham et al., 2013). The peak at 1388 cm⁻¹ corresponds to the bending vibrations of the C–H and C–O groups of the polysaccharides. The peak at 1155 cm⁻¹ was attributed to C–O–C asymmetric stretching vibration associated with CNCs₁ and CNCs₂. Other peaks include 2851 cm⁻¹ and 2920 cm⁻¹ that are assigned to the C-H stretching vibrations (Abu-Thabit et al., 2020).

It is necessary to note that the FTIR results should demonstrate cellulose components since CNCs originate mainly from them. For example, the peak at 897 cm⁻¹ corresponds to the C-O-C deformation and stretching in cellulose (Gonultas & Candan, 2018). In the current study, this band is found at 790 cm⁻¹, where it indicates that the original molecular structure of cellulose was still maintained even with the removal of lignin and hemicellulose components. Furthermore, the vibrations of adsorbed water molecules should be demonstrated at bands 1640 cm⁻¹ (Zheng et al., 2019; Haafiz et al., 2013; Mandal & Chakrabarty, 2011). This is due to cellulose and water molecules' strong interactions with each other. However, the band is not found at that specific wavenumber. The stretching vibration of the C–O–C bonds representing the cellulose 4-glycosidic links linkages of the d-glucose units are shown at peaks 1162 cm⁻¹ and 1034 cm⁻¹ in both samples. These peaks correspond to the typical cellulose structure shown in previous studies (Kubovský et al., 2020; Luzi et al., 2019). It can be concluded that the modification presented by sulfuric acid that was used to obtain CNCs is successful due to the significant peaks that represent the changes in its chemical structure. The CNCs were successfully extracted from methods (1 and 2) since most of the lignin and hemicellulose components were removed. Based on the FTIR spectrum, both CNCs samples had similar chemical structures and no significant changes were observed; therefore, any isolation methods (1 and 2) can be used in the future to attain cellulose nanocrystals. The absence of peaks around 1500 cm⁻¹ and 1600 cm⁻¹ indicates the removal of lignin and hemicellulose (Ng et al., 2021; Mazlita et al., 2016). These results are in agreement with the results of SEM analysis that indicates the removal of lignin and hemicellulose.

3.6 X-ray powder diffraction (XRD) analysis

Biomass of lignocellulosic components is composed of amorphous and crystalline regions according to Chandra & Madakka (2019). This amorphous region is present as a result of lignin and hemicellulose present in DP, on the other hand, the crystalline region is due to the presence of cellulose. Therefore, the treatment that is done chemically leads to the depolymerization of hemicellulose and the delignification of DPs, which tends to increase the crystallinity of cellulose obtained in the end. As shown in Fig. 5, the crystallinity nature of the fibers indicates the successful conversion of the extracted cellulose to nanocellulose. The XRD spectra of CNCs₁ are represented by diffraction peaks at 2 theta values 15.8°, 22.5°, and 33.5°. Similarly, the diffractrogram of the CNCs₂ from method 2 also shows similar peaks, which are corresponding to the lattice planes (Miller index) 110, 200, and 004, respectively (Guo et al., 2020; Zheng et al., 2019; Chandra et al., 2016). It is indicated from the obtained peaks of both samples that the structure of the cellulose was preserved and maintained. Moreover, it is indicated that utilizing H₂SO₄ did not alter the original cellulose crystal (Marett et al., 2017). The high crystallinity of the obtained CNCs from both methods is indicated by the sharp diffraction peak at 22.5° which was similar to the results obtained by Trilokesh & Uppuluri (2019) who found the sharp peak at 2θ = 22.6°. Since hydroxyl groups formed the intermolecular and intramolecular H-bonding, therefore, the crystalline arrangement order for each CNCs has been reported. The cellulosic chains’ movement is limited by the H-bonding which leads to their alignment next to each other forming the cellulose crystallinity (Chandra et al., 2016). From Eq. (4), the crystallinity index is calculated for CNCs₁ and CNCs₂ to measure if the % crystallinity changed under the two different methodologies. It was found that the apparent crystallinity of CNCs₁ was 69.99%, whereas the crystallinity of CNCs₂ was 67.79%. Furthermore, as mentioned previously, the % yield of CNCs₁ is slightly higher than CNCs₂, which could be attributed to the lower cellulose content and lower crystallinity. In general, there were no significant differences in the position of the XRD peaks between the CNCs₁ and CNCs₂. Furthermore, the consecutive chemical treatments of DP effectively removed the non-crystalline regions which caused the high crystallinity. These results support the FTIR and SEM results in terms of lignin and hemicellulose removal and confirm that the obtained product is CNCs. Table 4 represents the crystallinity percentage of our obtained CNCs and other CNCs obtained from different resources by acid hydrolysis.

Close

Fig. 5

XRD patterns of CNCs₁ and CNCs₂.

Export to PPT

3.7 Cost analysis, research implications, and limitations

Cost plays a crucial role in choosing the optimum CNCs extraction method, hence, is considered as an important driver for the success of the extraction process. Generally, assuming that the CNCs source is cost-effective when it does not demand high processing and it is easily available in nature, an industrial by-product, or waste material (Yaqub et al., 2014). Since the fiber business and its products reached around USD 1.08 billion in the global market, various resources have been used to produce nanocellulose. Agro-waste residues are a sustainable source of cellulosic biomass such as fruit waste, wheat straw, and date pits, in which tremendous efforts were applied for mining cellulosic fibers from such wastes (Yang et al., 2017). The cost of the extraction process can be reduced not only by using the non-valuable agricultural wastes to produce highly profitable products such as nanocellulose but also by the pretreatment process that can be performed before nanocellulose extraction. Moreover, Visanko et al., (2014) mentioned that using treatment processes such as alkaline treatment and acid hydrolysis for the production of nanocellulose instead of the highly intensive machine-driven process would reduce the required energy from 30,000 kWh/ton to 1000 kWh/ton. Based on our knowledge and the research outputs, using the Soxhlet apparatus technique (method 2) for extraction of CNCs from DP was more time consuming and energy-intensive since it required more time on the hot plate compared to the mechanical stirrer technique (method 1) as mentioned previously in the methodology. Moreover, using the Soxhlet apparatus (method 2) demanded more chemicals than the mechanical stirrer (method 1) since hexane was used only for the Soxhlet technique which increases the final cost of the extraction process. Therefore, using the mechanical stirrer technique is more cost-effective compared to the Soxhlet apparatus method.

Furthermore, sulfuric acid is commonly used for the acid hydrolysis process due to its low cost and better stability of the CNCs suspensions. However, discharging large amounts of the acid and a long time for washing the obtained CNCs is one drawback of using such acid. Considering the environmental issues caused by discharging wastewater with strong acids like sulfuric and hydrochloric acid encourages finding alternatives with less environmental impact. For this reason, phosphoric acid has been used as a promising alternative for sulfuric acid in acid hydrolysis as it showed good thermal stability, but more research is required to increase its yield, which will eventually affect the final cost of the process (Gao et al., 2020; Tang et al., 2015). The presence of hemicellulose, lignin, and other undesired organic materials present a challenge in the production of CNCs, but this can be overcome by pretreatments such as bleaching and alkaline washing which are able to remove all these materials from the biomass. Since challenges do exist, more efficient research outputs and environmentally friendly processes are required to overcome these issues and to extract CNCs in a sustainable way.

Cost analysis for both methods was proposed by including important chemicals and materials used, energy consumption, and other expense as shown in Table 5. The cost breakdown calculations were based on 50 g of DP and the percent yields of CNCs₁ and CNCs₂ were 12% (i.e. 6 g) and 10% (i.e. 5 g). The yield is calculated and mentioned in the manuscript in Eq. (3). It was concluded that method 1 is less expensive than method 2 based on the breakdown of the cost of each step for CNCs production. Keep in mind that each extraction and isolation step of CNCs from both methods could have a potential concern regarding its sustainability towards the environment. However, future perspectives and integrated processes towards a cleaner and more sustainable approach can be achieved. For example, the dewaxing step in method 2 can be used to produce biodiesel and the black lignin solution can be used for the generation of energy (Abu-Thabit et al. 2020). Therefore, by-products produced throughout the entire production stage can be used for other purposes if the concept of sustainability is taken into consideration. This can provide a way to properly utilize residue and waste into value-added products.