Effects of Hemicellulose on Recycling Performance of Paper Based on Sisal Fibers

The pulp and paper industry growingly paid attention to the recycling and maintenance of waste paper products. Each paper-making cycle would lead to a sharp drop in the mechanical properties of the cellulosic paper, which was related to the hornication effect. Here, the recycling performance of the holocellulose paper was studied, compared with that of the cellulosic paper. Holocellulose bers from sisal were fabricated by a gentle delignication method, and the well-preserved cellulose and hemicellulose components hindered the cocrystallization and aggregation of cellulose bril. Holocellulose paper exhibited much more favorable recycling properties, compared with cellulosic paper. After 5 runs of recycling, holocellulose paper still shown an ultimate strength as high as 25 MPa (reduced from 35 MPa), a decrease of 27.1 %. However, cellulosic paper experienced a substantial loss in ultimate strength from 35 MPa to 9 MPa, a decrease of about 74 %. This can be attributed to the core-shell structure from cellulose and hemicellulose to weaken the hornication effect.


Introduction
The disposal of waste paper had always attracted much attention in the paper industry (Luo et al., 2003).
Simply discarding it for natural degradation or incineration would cause a waste of biomass resources. Meanwhile, it was likely to cause environmental pollution and unexpected natural accidents, such as soil pollution and re disaster (Ouadi et al., 2019). Therefore, an environmentally friendly strategy to utilize waste paper and produce new recycled paper was of importance (Jeong et al., 2018). It could reduce the burden of environmental pollution and improve economic bene ts (Lin et al., 2020). For instance, the recycling rate of waste paper in China had reached a high level (70%) in 2017. Meanwhile, the recycling of waste paper was an effective way to reduce carbon emissions in China's paper industry (Shang et al., 2021). However, the mechanical strength of the paper would drop to a frustrating level during the recycling process, compared with the original paper ber (Janda, 1995). Fiber qualities would undergo an irreversible decline when recycled paper ber was exposed to cycles of wetting and drying. In detail, the morphology of the cellulosic ber had changed during the recycling process (Mendes et al., 2019), bers trended to aggregate and formed a more compact structure (Ponni et al., 2012). The amorphous area of the ber was more easily degraded, and relatively, the crystallinity became larger, which made it more di cult to disperse into a single ber. The term "horni cation" aptly described this phenomenon, the degree to which bers were irreversibly combined when they were dried or rewet (Cid et al., 2020;Mendes et al., 2019;Yang and Berglund, 2020). High-strength beating was intended to disperse the bers as much as possible (Chen et al., 2012). It was the most commonly used method to overcome the effects of horni cation (Suchy et al., 2010). Furthermore, the addition of ber binder or fresh ber can also improve the mechanical properties of recycled paper (Santmarti et al., 2020;Tavast et al., 2014;Yang and Pelton, 2017). However, the recycled paper was not able to fully recovered to the performance of the original paper.
Hemicellulose, as a substance without crystalline regions, was tightly wrapped on the surface of cellulose(Terrett and Dupree, 2019). It could be intimately integrated into cellulose bers and hinder the horni cation of cellulose paper during recycling processes (Déléris and Barroca et al., 2001), chlorine bleaching method (Axegård, 2019), and sodium chlorite method (Jin et al., 2019). The sodium chlorite method was called the Wise method that the purpose of continuous deligni cation was achieved by circulating added sodium chlorite and glacial acetic acid (Hubbell and Ragauskas, 2010). The sodium chlorite method was one of the most common methods to selectively remove lignin in industry scale. The main product in this reaction was chlorine dioxide, and chloride and chlorate, as the secondary products (Siqueira et al., 2013). It took into account a high selectivity and low environmental impact. Therefore, the sodium chlorite method was still a commonly used method in laboratories.
These hemicellulose-rich holocellulose bers would be suitable for the investigation of the recycling potential. Moreover, it was interesting to compare the recycling performance between holocellulose paper and cellulose paper from the same raw material. In this study, holocellulose paper was prepared via the sodium chlorite method, followed by the de bering and paper-making process. As a comparison, an extra alkali treatment was carried out to remove hemicellulose from holocellulose bers, and thereby achieve cellulose paper. Then, the recycling performance of two kinds of paper was investigated comprehensively.

Preparation of sisal holocellulose and cellulosic ber
First, sisal leaves were cut into a rectangle shape with 1 cm *1.5 cm size, then washed and dried. 50 g of dried sisal leaves were soaked in 500 mL acetone overnight to dewax, and then washed repeatedly with deionized water and ethanol until the acetone was completely removed. Then, sisal leaves (10 g) after dewax were added into the aqueous solution (350 mL) with sodium chlorite (3.75 g) and glacial acetic acid (3 mL) for 1.0 h at 75 °C. The same amount of sodium chlorite and glacial acetic acid was added every hour for 4 times. Finally, the resultant product was repeatedly washed by running water to remove the residual chemicals. In order to beat the product into ne holocellulose ber, the product was added into a Vitamix Blenders (20000 rpm, TNC5200, USA) for 3 minutes. Cellulosic bers were obtained by adding holocellulose bers into the NaOH aqueous solution at 90 °C for 2.0 h.

Paper making and recycling process
A Rapid Köthen (RK3AKWT, Austria) was carried out for paper making. The paper with a targeting weight per surface area at 165 g/m 2 was dried at 90 °C and 0.1 MPa for 10 min. During the recycling process, the holocellulose paper was torn into small pieces and soaked in water. Then, it was dispersed into a slurry used a ber uffer at 5000 revolutions and made into paper again. A total of 10 times were recycled, and the obtained papers with different recycling times was named holocellulose-0~10. The cellulose paper was recycled 5 times in the same way and named cellulose-0~5.

Characterizations
The acid-insoluble lignin content of sisal holocellulose was determined through the GB-T2677.8-1994 standard. The obtained monosaccharide solution was adjusted to a concentration of 1000 ppm. The content of cellulose and hemicellulose was measured by a high pressure liquid chromatography (LC-100, USA) with a differential detector. A ber analyzer (FS300, Finland) was performed to measure the length and width of the ber. A universal material testing machine (INSTRON 5565, USA) was carried out to test the mechanical tensile properties of the holocellulose paper, and each paper was tested three times. Fourier transform infrared spectrometer-infrared (FT-IR) microscope (Nicolet IS50-Nicolet Continuum, USA) was recorded to test the groups of recycled paper and the self-reinforcing paper. A multi-position automatic sampling X-ray diffractometer (XRD, X'pert Powder, PANalytical) was used to determine the crystallinity of the holocellulose paper with a conventional wide-angle from 5 ° to 60 ° at a scanning speed of 12 °/min. The crystallinity index (CrI) was calculated by the peak strength method as follows: (Segal et al., 1959) Note: I 200 was the diffracted intensity of crystallographic area at 2θ ≈ 22.5°. I am was the diffracted intensity of the amorphous region at 2θ ≈ 18.8 °.
The holocellulose paper with a size of 6 cm *10 cm was carried out by a 0-spacing tensile strength tester (Z-span 2400, Canada) to test the 0-span tensile strength.
The water retention values (WRV) of recycled paper were calculated by the ISO 23714:2014 standard. In detail, about 1 g of pulp was centrifuged at 3000 g for 30 min to remove water and weighed. Then, the resultant pulp was dried at 105 °C for 24 hours and weighed again. The following was the calculation formula for WRV: Note: m wet was the mass of the pulp before drying, and m dry was the dry weight of the pulp.
The following was the formula for calculating the degree of horni cation (DOH) Note: WRV 0 was the water retention value of holocellulose-0 and WRV n was the water retention value of holocellulose-n.
The surface morphology and ber morphology of the tensile fracture surface was observed by a eld emission scanning electron microscope (SEM) (EVO 18, Germany).
Results And Discussion Figure 1 described the entire paper making and recycling process. The raw material extraction process was divided into two steps. Firstly, the holocellulose was extracted by gentle deligni cation, and then the hemicellulose in the holocellulose was destroyed by alkali treatment. Holocellulose bers and cellulose bers were obtained by the above two steps. During the typical paper making process, the holocellulose/cellulose was de bered and then ltered through the vacuum to obtain wet cake. The Kaiser system can simply and quickly remove the moisture to obtain a dense paper. During the recycling process, the paper was torn into small pieces and then placed in the ber decomposer. The strong force would redistribute the small pieces of paper into a free ber state so that the paper could be made again.
In each cycle process, changes in ber morphology and recycled paper properties would occur. It was of importance and meaningful to explore the evolution of the morphology and recycling performance of holocellulose paper during the recycling process.
With the same targeting weight per surface area at 165 g/m 2 , holocellulose paper with a smooth surface ( Fig. 1B) possessed higher thickness (240 µm) and lower density (0.69 g/ cm 3 ), compared to that of cellulosic paper (230 µm, 0.76 g/ cm 3 ) (Table 1). Native hemicellulose would expand the inter brillar space between cellulose bers during the paper making process, resulting in lower density and higher porosity of holocellulose paper (Yang and Berglund, 2020). As SEM images shown in Fig. 1B, Fibers in holocellulose paper appear similar in structure compared with cellulosic paper in size and distribution ( Fig. 1B and 1C).
As shown in Fig. 2A and Table 1, the ultimate strength of holocellulose was close to that of cellulosic paper, supported by the SEM results in Fig. 1B and 1C. Moreover, compared with cellulosic paper, holocellulose paper exhibited a lack of strain-softening, due to the favorable inter ber bonding in holocellulose paper with native hemicellulose. Upon recycling, the mechanical properties of cellulosic paper decreased signi cantly, which was attributed to the horni cation effect of cellulose bers. In contrast, holocellulose paper showed no apparent changes in the mechanical performance (Fig. 2B). For example, as shown in Fig. 2C, the ultimate tensile strength (σ u ) of holocellulose paper was higher than that of cellulosic paper during the recycling process. After 5 runs of recycling, holocellulose paper still shown an ultimate strength as high as 25 MPa (a decrease of 27.1%), while cellulosic paper experienced a substantial loss in ultimate strength from 35 MPa to 9 MPa, a decrease of about 74%. Meanwhile, the strain to failure of cellulose paper showed rapidly dropped after 2 runs of recycling from 4.28 to 2.50, a decrease of 41.6% (Fig. 2D), while this phenomenon did not appear in the recycling process of holocellulose paper. Similarly, the Young's modulus of elasticity (E) calculated from Fig. 2A and Fig. 2B exhibited favorable performance of holocellulose paper, as shown in Fig. 2E. After 8 runs of recycling, holocellulose paper presented a Young's modulus of elasticity as high as 1.1 GPa (a decrease of 17%).
Moreover, after 5 runs of recycling, the 0-span tensile strength of holocellulose paper showed not changed (115 N/cm), while the cellulosic paper displayed an obvious decrease ( Fig. 2F and Table 1). Consequently, holocellulose paper showed favorable recycling performance related to the mechanical properties compared with cellulosic paper.
The content of cellulose and hemicellulose was measured by high pressure liquid chromatography. Although holocellulose ber was obtained by the deligni cation process, as showed in Fig. 3A and Table   S1, holocellulose ber was consisted of cellulose (~ 77%), hemicellulose (~ 10%), and lignin (~ 7.5%). Hemicellulose was destroyed by treated with an alkali solution to obtain cellulose(Carvalho et al., 2019; Zhang et al., 2020). Upon recycling process, the content of hemicellulose slightly decreased, indicating a small amount loss of hemicellulose during the paper making procedure. Thus, this result strongly supported that the presence of hemicellulose was conducive to the recycling performance of holocellulose paper compared with cellulosic paper. To further clarify how the native hemicellulose affects the recycling performance, the FT-IR spectra of holocellulose and cellulosic paper with different runs of recycling were studied in Figure 3B. There was a clear difference between holocellulose and cellulosic paper in terms of hydroxyl characteristic peaks. The hydroxyl characteristic peaks gradually increased with the recycling processes of cellulose paper. As expected, this suggested that cellulose ber trended to aggregate and formed a more compact structure via irreversible hydrogen bonds during recycling processes, which was consistent with the decreased water retention value (WRV) and increased degree of horni cation (DOH) displayed in Figure 3C and 3D. The reason for the degradation of recycling performance was attributed to the horni cation effect between the bers, which caused the bers to be unable to reach the dispersion degree of the previous papermaking in the process of rede ating. The change in water retention value was an important criterion that re ected the degree of horni cation. The higher the water retention value, the fewer irreversible hydrogen bonds between bers, and the better its dispersibility.
Moreover, during the recycling process, the crystallinity of cellulose calculated from XRD spectra ( Figure  S1) increased signi cantly after different runs of recycling. In contrast, the small change of hydroxyl characteristic peak in Figure 3B and low DOH values shown in Table 2 jointly explained that hemicellulose in holocellulose paper could prevent the horni cation effect. Combined with the above result, the slight changes of hemicellulose content ( Figure 3A) during the recycling process could prevent the horni cation effect and maintained the mechanical performance of holocellulose paper. Furthermore, the SEM images could also explain the changes in the holocellulose and cellulose ber structure at different cycle stages (Figure 4). All broken bers in the image are from tensile tests. Interestingly, the tensile fracture surface of holocellulose-0 and holocllulose-5 bers appeared at, while cellulose-0 and cellulose-5 bers appeared a larger deviation, indicating that the strong inter ber bonding was formed in holocellulose paper.

Conclusion
Through the gentle deligni cation method, the hemicellulose-rich holocellulose was prepared from sisal raw material. The recycling process of waste paper was simulated, compared with the performance changes of holocellulose paper and cellulose paper. After 5 times of recycling, the ultimate tensile strength of the holocellulose paper decreased by 27.1%, and after 10 times of recycling, the ultimate tensile strength decreased by 57.5%. The cellulose paper was recycled 5 times and it dropped by ~ 74%. It can be seen that the hemicellulose-rich holocellulose paper presented higher recycling performance than cellulose. From the ber point of view, both holocellulose and cellulose, recycling processes would cause a trend of decreasing ber width. The hemicellulose in the holocellulose assumes a sacri cial role, reducing the damage suffered by the cellulose ber during the recycling process. As a consequence, holocellulose had shown signi cant advantages in recycling processes, the ber morphology would not change much during the recycling processes. The recyclability of bers was greatly increased. Recycled paper was more competitive in terms of market demand. Figure 1 (A) Schematic diagram of raw material extraction from sisal to recycle paper. Optical view and SEM images of (B) holocellulose-0 paper, and (C) cellulose-0 paper.