Cellulose from Tamarix aphylla’s stem via acetocell for cadmium adsorption

2.1 Tamarix aphylla's composition

According to the method reported by the Technical Association of Pulp and Paper Industry (TAPPI), the chemical composition of Tamarix aphylla is determined. The insoluble lignin is determined according to the Klason method (T222 om-06). The percentage of soluble compounds in both hot water and cold water, and 1% sodium hydroxide aqueous solution and ethanol-toluene mixture were determined according to the TAPPI standard methods (T207 cm-08, T212 om-07, T204 cm-07). Then, following to the standard method (T211 om-07), the ash content was determined by calcination of Tamarix aphylla’s stem at 525 ± 25 °C for 4 h.

2.2 Cellulose obtainment

The acetocell process was followed to pretreated Tamarix aphylla stem’s powder with acetic acid. The biomass powder was transferred with an aqueous acetic acid into a round bottom flask. Different samples were synthesized by varying the concentration of this solution; 20, 40, 80 mg L⁻¹, with the treatment time extent between 2 h and 8 h. Then the mixture was filtered and washed with distilled water (Marques et al., 2021).

The previous pretreated sample has undergone a bleaching process by adding an aqueous sodium chlorite solution 1.7% (w/v), was assisted by a buffer solution (2.7 g NaOH in 50 mL of water with 7.5 mL of acetic acid and the addition of water to a volume of 100 mL). The liquid ratio NaClO₂/buffer acetic solution/H₂O was 1:1:3 (Nabili et al., 2016).

The last step of the treatment was the differential solubilization, which was achieved according to the standard method TAPPI (T203 cm-99) using 15 mL sodium hydroxide (17.5%) at 25 °C for 40 min with stirring. The cellulose obtained was filtered and washed vigorously with distilled water until a neutral pH was reached.

2.3 Characterization

Fourier transform infrared spectra were recorded using a Perkin-Elmer FTIR spectrophotometer between 400 and 4000 cm⁻¹.

The X-ray diffraction analysis was performed on DS and CDS powder. The samples were placed in a 2.5 mm deep cell and the measurements were performed with a PANalytical, X’Pert PRO MPD diffractometer (PANalytical, Massachusetts) equipped with an X’celerator detector.

The elemental distribution was evaluated through SEM assisted by energy dispersive EDX in a Joel JSM 400 equipment with a silicon drift detector. Scanning electron microscopy (SEM) (model Philips 30) was used to observe the surface morphology of the treated fibers in each stage of treatment.

Florescence optic microscope where performed using the Axio. ImagerM1m, the pictures were obtained by a high resolution camera 3-CCD (1360x1024 pixels) JVC KY-F75U and it were extracted through Axion Vision Relase 4.8 software.

The analyzer MORFI LB-01 (Techpap, France) was used to determine the suspended fibers (weighted length, width and of fine elements).

The thermal stability of the samples was determined by a thermo-gravimetric analyzer STA 6000 (Perkin Elmer Instruments, Buckinghamshire), which measures the weight loss of the samples during the temperature’s increase.

The cellulose’s degree polymerization was determined from the intrinsic viscosities [η] of cellulose solutions in cupri-ethylenediamine/water at 25 °C.

The Brunauer–Emmett–Teller surface area (SBET) and pore structure parameters of the cellulose were obtained from nitrogen adsorption desorption measurements at 77 K using a Micromeritics ASAP 2020 instrument.

2.4 Batch adsorption experiments

Initial cadmium aqueous solution was prepared 100 mg of cadmium chloride per 1L of deionized water. Then various concentrations from 10 to 45 mg L⁻¹ were gotten from the initial solution through dilution. Various samples were prepared by varying different parameters.

The removal of cadmium ions from an aqueous solution onto prepared cellulose was carried out by batch method. After the adsorption experiment, the equilibrium adsorption capacity Q_ads (mg g⁻¹) was calculated according to equation below

For the absorbent amount evaluation on the adsorption procedure, assays were performed at the same temperature and contact time used in the previous experiments with conserving the (pH = 6) except we varied the adsorbent dose; 10, 15, 20, 25, 30, 35, 40 and 45 mg.

To test the effect of the pH of the solution, the adsorption process was carried out at 25 °C using 10 mg of biosorbent with 10 mL of the aqueous solution contaminated with cadmium (45 mg L⁻¹), the pH was adjusted from 2 to 12 using HCl and NaOH solutions (0.1 mol L⁻¹). The different solution samples were subjected to magnetic stirring for 1200 min. Then, the samples were filtered and cadmium ion concentration was determined by Flame Atomic Adsorption Spectroscopy (FAAS) Fisher Scientific iCE 3000.

In purpose of evaluation of concentration of cadmium, ions Cd²⁺ on aqueous solution effect, various concentrations from 10 to 45 mg L⁻¹ with respecting the followed conditions earlier as cellulose amount (10 mg), contact time (1200 min), pH = 8.

In order to describe the kinetic of the cadmium adsorption onto cellulose, time extent from 60 to 1200 min were evaluated in this step while retaining the used previously conditions with cadmium ions concentration of 45 mg L⁻¹.

2.5 Adsorption isotherms and kinetic models

The Freundlich, Langmuir, Dubinin– Radushkevich and Temkin isotherms were followed to evaluate the experimental points for cadmium adsorption (Table 1). The pseudo-first-order (Eq. (2)) and pseudo-second (Eq. (3)) order model were used for kinetic adsorption estimation (Ho and McKay, 1999; Lagergren, 1898; Saad et al., 2014).

3.1 Tamarix aphylla’s composition

The chemical composition of Tamarix aphylla and other plants from the literature were summarized on Table 2. The percentage of the isolated cellulose from Tamarix aphylla (39.0%) was similar to that of Mechi’s work (Mechi et al., 2017) and comparable to that of Nitraria retusa (41.0%) (Ferhi et al., 2014a), moreover, found to be lower than cellulose content for some annual plants namely Pituranthos chloranthus (46.6%) (Ferhi et al., 2014a), Stipagrostis pungens (44.0%) (Ferhi et al., 2014b) and Opuntia ficus-indica (53.6%) (Mannai et al., 2016). The ash’s percentage was around 3.5%, which is lower than that presented by Mechi’s work (10%) (Mechi et al., 2017). However, it was similar to other plants such as Retama raetam (Ferhi et al., 2014a) and Astragalus aramtus (Moussaoui et al., 2011). In contrast, Tamarix aphylla has an important lignin percentage of 30%, which was higher than that of annual plants listed in Table 2 and particularly for that of Opuntia ficus-indica, which not exceeds 4.8% (Mannai et al., 2016). This difference on composition distribution from one biomass to another can be explained by the variance of botanical source of the raw material (Sukumar et al., 1991).

Tamarix aphylla was rich in mineral compounds (Fig. 1, Table 3). The calcium and sulfur were present with a significant mass ratio on Tamarix aphylla, which were respectively 14% and 17%. The swamp sol was rich with sulfur (Driese et al., 2021), thus it provide an important amount of this last to Tamarix aphylla, in which it was grown. Other elements such as sodium, magnesium and potassium were present with modest percentages, moreover traces of aluminum and copper were observed as well.

Close

Fig. 1

SEM-EDX spectra of Tamarix aphylla’s ash.

Export to PPT

3.2 Effect of temperature on delignification efficiency

Generally, temperature is a kinetic factor that speed the chemical reaction. Indeed, the diagram of Fig. 2 illustrates the effect of temperature on cellulose bleaching. The rise up of temperature to 100 °C during the bleaching allowed the decomposition of sodium chlorite NaClO₂ into ClO^-, which reacted subsequently with the aromatic ring of lignin to destroy it. Thus, the split-up of the ether bonds that linked hemicelluloses and lignin was taken place resulting on the elimination of non-cellulosic compounds (Hornsby et al., 1997; Yue et al., 2015). At 100 °C, the bleaching lasted eight hours. However, at 75 °C, the reaction requires more than ten hours to get the pure white color of holocellulose and more than 24 h with a lower temperature of 50 °C.

Close

Fig. 2

Effect of temperature on bleaching process.

Export to PPT

3.3 Characterization and chemical proprieties of cellulose

The color of the raw material is brownish-yellow then changes from brown after acetocell pretreatment to pure white cellulose after bleaching. This color change is due to the removal of non-cellulosic compounds mostly lignin. The pure white color of the final fiber clearly shows the success of the delignification process. The cross-section of Tamarix aphylla’s stem (Fig. 3.a) shows a homogeneous morphology on the surface, a tegument shape, it was made from several cellular strads arranged perpendicular to the albumen cells. The longitudinal section (Fig. 3.b) shows clearly that the albumen consists of hexagonal oblong cells juxtaposed of 30 μm with thick walls and well-built. At the cross-sectional level, short, tight cells have the shape of a bee’s beehive (Fig. 3.c); clear pores are observed on the surface of raw material (Fig. 3.d).

Close

Fig. 3

Scanning electron micrographs of raw material shape: Transversal section (a), tegument (b), albumen (c), longitudinal section (d), pretreated fiber (e), holocellulose (f) and cellulose (g).

Export to PPT

During the delignification process, some morphological changes have been observed owing to the exclusion of non cellulosic fibers (Fig. 3.e, f, g). After acetocell pretreatment, the fiber surface becomes detached, which indicates the partial removal of the non-cellulosic materials such as lignin. This elimination cause some morphological changes for cellulose, such as broken ends that its percentage is around 18.62% (Table 4). According to previous reports (Hornsby et al., 1997; Yue et al., 2015), lignin forms a bridge bond with the cellulose bond so it acts as a binder in the fiber components. Hence, the preserving of the organized form after delignification treatment cannot be possible and the destruction of the combined structure is a predictable consequence. Indeed, the disjointing of the cellulose fibers which appear and became more and more obvious after each step of treatments due to the progressive removal of cemented material (Yue et al., 2015). The formation of cellulose fibers was not newly produced but they were existed but they mainly appeared after gotten ride off the lignin and extractible and other compounds of Tamarix aphylla.

The separation of the cellulose fiber was confirmed by the Florescence optical microscopy observation (Fig. 4) and it was accompanied by a diameter reduction from raw material (32 µm) to cellulose (21 µm). This size reduction was mainly attributed to the elimination of lignin and hemicelluloses (Yue et al., 2015), and it causes, as a result, an increase in the fiber aspect ratio (length/diameter), which improve the reinforcing ability of the fibers for composite applications.

Close

Fig. 4

Florescence optical micrographsof cellulose.

Export to PPT

The cellulose polymerization degree was estimated at 598 according to Mark-Houwink equation (Kasaai, 2007; Paul and Lodge, 2007). This value was in an average of cellulose degree of polymerization’s scale, it was higher than the polymerization degree of cellulose extracted from Posidonia oceanic, which was 153 (Khiari et al., 2010), and it is lower than the fiber extracted from annual plants namely Nitraria retusa, Pituranthos chloranthus, Stipagrostis pungens, Retama raetam, Astragalus aramtus and Opuntia ficus-indica (Ferhi et al., 2014a; 2014b; Mannai et al., 2016), with a polymerization degree higher than 1000. In addition the Specific surface area of the isolated cellulose was around 1 m² g⁻¹, which almost matching to that of cotton cellulose film (0.72 m² g⁻¹) reported by Rowen and Blaine (1947) and to that described one by Emmett and Dewitt (0.60 m² g⁻¹) (Emmett and Dewitt, 1941). Nevertheless, it is much lower than other biosorbent such as activated carbon from date palm petiole which was around 655 m² g⁻¹ (Khadhri et al., 2019).

The obtained infrared spectra of raw material, pretreated, bleached sample, and cellulose were illustrated in Fig. 5. The band near 3400 cm⁻¹, observed in all spectra, was attributed to stretching vibrations of a hydroxyl group (O-H). The raw and pretreated sample shows a band around 1720 cm⁻¹ relative to the C⚌O bonds of the carbonyl group which are corresponding simultaneously to the ester groups in hemicellulose and carboxylic acid groups in lignin (Hornsby et al., 1997; Yue et al., 2015). The disappearance of this band in the spectrogram (Fig. 5.d) confirms the full missing of C⚌O groups in the obtained cellulose so its purity from any trace of other polysaccharides (Alemdar and Sain, 2008). All spectra present a band at 1640 cm⁻¹ that corresponds to the absorbed water. The band around 1500 cm⁻¹ assigned to the lignin aromatic ring, was present in the raw material spectra and disappeared in both holocellulose and cellulose (Fig. 5) (Sun et al., 2005). The band in 890 cm⁻¹ observed in all spectra refers to stretching vibration of C-O-C assigned to glycosidic bridges of cellulose (Alemdar and Sain, 2008).

Close

Fig. 5

FTIR spectrogram of raw material (a), pretreated fiber (b), holocellulose (c) and cellulose (d).

Export to PPT

The crystallinity index (CrI) of the different samples was determined based on the reflected intensity data following the method reported by Segal et al. (1959).

The results reveal an increase in the crystallinity index through the chemical treatment of Tamarix aphylla’s stem (Fig. 6). The cellulose crystallinity index (52%) was higher compared to the holocellulose, pretreated and untreated fibers, which were respectively 48%, 42% and 32%. After each treatment, the crystallinity index of sample increase due to the remove of amorphous compounds such a hemicelluloses and lignin. These observations are accorded to those of Alemdar and Sain (2008) and Sun et al. (2005) whose work has shown that after chemical treatment the crystallinity index of fiber must be definitely increase.

Close

Fig. 6

X-ray diffractogram of raw material (a), pretreated fiber (b), holocellulose (c) and cellulose (d).

Export to PPT

The TGA curves (Fig. 7) show some common masses' losses and different other ones. The decomposition starts slightly later especially in the case of raw material and pretreated sample. All TGA curves show a slight common mass loss at 100 °C. It is all obvious an evaporation of moisture; corresponding to the fiber dehydration by releasing the absorbed water. At temperature range 250–310 °C, a significant mass loss corresponding to the hemicelluloses decomposition. The decomposition of hemicelluloses has decreased after each step of treatment from the raw material 30% until the cellulose 10%. The hemicellulose was progressively eliminated after treatments. Due to its quasi-crystalline and cohesive structure, cellulose requires a high temperature for the breakdown of glycosidic bonds and hydrogen bonds. Therefore, its decomposition was observed at a temperature; from 330 to 400 °C. The cellulose curve shows a quasi total mass loss at this range of temperature (Alemdar and Sain, 2008; Hornsby et al., 1997; Sun et al., 2005; Yue et al., 2015).

Close

Fig. 7

TGA curves of raw material (a), pretreated fiber (b), holocellulose (c) and cellulose (d).

Export to PPT

The lignin’s aromatic rings started decomposing at 200 °C and continued to 600 °C since it has required an elevated temperature to break the strong aromatic ring bonds. The TGA curves of the untreated (Fig. 7.a) and treated (Fig. 7.b) samples show mass losses of 30% and 20%, respectively, at 200 °C. Contrary, there is no mass degradation at that specific temperature of lignin decomposition (600 °C) in both holocellulose and cellulose curves (Fig. 7.c and Fig. 7.d). This confirms the absence of lignin so to the effectiveness of the delignification process “acetocell” and the followed step of treatment to cellulose fiber isolation.

3.4 Adsorption outcomes

Within increasing the pH, tending from acidic to alkaline one, the capacity of Cd²⁺ adsorption onto cellulose is increasing. The influence of pH on cadmium adsorption capacity was significantly presented 3 levels in the Fig. 8.a. From 2 to 4 pH, the adsorption capacity recorded its lowest values which were respectively 24.50 and 28.01 mg g⁻¹. Up to 6 pH, the Cd²⁺ adsorption capacity was more promising with 42.21 mg g⁻¹, and it remain increasing to reached the peak with 44.21 mg g⁻¹ at 8 pH. Nevertheless, the adsorption capacity was decreased to 35 mg g⁻¹ at pH 12. The cadmium uptake decreased at 10–12 pH range but it is still higher than at acid solution (Ozel, 2012). In fact, the pH_PZC of cellulose was 4.8, at pH below than pH_PZC, the surface of cellulose was positively charged. Thus, the Cd²⁺ uptake was inhibited due to the possibility of the competition between H⁺ ions and cadmium on the surface of the cellulose at low pH values. This is explanation was admitted by earlier works of Ozel (2012) and Zamani et al. (2013). However, at pH above pH_PZC, the cellulose surface is negatively charged. Consequently, the Cd²⁺ was well adsorbed by the O⁻ ions of cellulose at pH > pH_PZC. Indeed, the adsorption was very privileged at pH 6–8, this is justified by the natural of the electrostatic interaction link between Cd²⁺ and cellulose O⁻ which enhanced the adsorption process. The obtained result was accorded to earlier work on the efficiency of adsorption process of heavy metal at alkaline medium especially around this range of pH (Ozel, 2012; Udoetok et al., 2016; Zamani et al., 2013).

Close

Fig. 8

Effect of pH (a), cellulose amount (b), time (c) and cadmium concentration (d) on adsorption of cadmium ions onto cellulose.

Export to PPT

Once we increase the cellulose amount from 10 mg to 45 mg, the elimination rate of Cd²⁺ increasing toward reach the highest value (0.983) (Fig. 8.b). This last almost equal to 1 which present the total of cadmium’s elimination rate. Axiomatic, more cellulose quantities along the adsorption process provide extra liber O⁻ ions of cellulose which improve the creation of chemical interactions with the Cd²⁺ ions, which lead then for their removal subsequently the cadmium uptake increasing (Gueye et al., 2014; Khosravi et al., 2014; 2018; Wang et al., 2010). These results are coherent with those obtained by Khosravi et al. (2014) and Gueye et al. (2014).

Since we have worked at room temperature (25 °C), the cadmium uptake was increasing progressively with more adsorption process time extent (Fig. 8.c). We needed the largest period of time; 1200 min to reach the maximum of cadmium’s elimination rate (0.978). Therefore, we fix this temperature for the rest of the experiment steps.

Moreover, the cadmium ions concentration influence on its uptake on cellulose. The adsorption capacity increases proportionally with cadmium concentration (Fig. 8.d). Best cadmium uptake recorded with the highest concentration value 45 mg L⁻¹. More cadmium ions released on the treated solution enhance the chemical links creation with the cellulose ions which improve subsequently the removal of cadmium (Wang et al., 2010). This finding is in line with the results of the study of Khosravi et al. (2014).

Adsorption capacities of sundry adsorbents, cellulosic and non cellulosic ones were illustrated in the Table 5. Comparing with others adsorbent, the cellulose from Tamarix aphylla reported in the present study displayed a significant cadmium adsorption capacity 44.21 mg g⁻¹ at pH 8 which was higher value than those obtained using mesoporous silica and cellulose/chitin beads (20 wt% chitin) which were only 3.62 mg g⁻¹ (Lalchhingpuii et al., 2017) and 35.9 mg g⁻¹ (Zhou et al., 2021), respectively. Nevertheless, Zeolite-supported nanoscale zerovalent iron (Z-NZVI) (Li et al., 2018), Zeolite-supported microscale zerovalent iron (Z-mZVI) (Kong et al., 2017), HMO-P(HMAm/HEA) hydrogel (Zhu and Li, 2015) and orange peel (Tran et al., 2016) were revealed Q_max values of 62.02 mg g⁻¹, 63.14 mg g⁻¹, 93.86 mg g⁻¹, and 59.5 mg g⁻¹,respectively, which were higher than that of cellulose from Tamarix aphylla. However, comparing with raw α-cellulose, the surface-modified cellulose and Sulfur–chelating cellulose, the cellulose presented in this work have a Q_max much higher than that correspond to α-cellulose which was around 7.47 mg g⁻¹ (Widyayanti et al., 2021) but it was slightly lower than that of Sulfur–chelating cellulose (54.71 mg g⁻¹) (Zheng et al., 2019). The cellulose obtained from Tamarix aphylla was privileged for cadmium adsorption than modified celluloses from pineapple (Cell –EDTA and Cell-CM) (Daochalermwong et al., 2020), which provide only a value of Q_max of 33.2 mg g⁻¹ and 23 mg g⁻¹, respectively.

To sum up all the outcomes stated above, we can conclude that unmodified cellulose from Tamarix aphylla, exhibited a significant capacity for cadmium adsorption and better than other celluloses, this is mainly the originality approach of our work.

3.5 Kinetic study and adsorption isotherms

Origin 6.0 software was used to estimate kinetics and equilibrium parameters using nonlinear regression 6.0 software. The kinetic study was performed through pseudo-first-order and pseudo-second order model, which were calculated using the previous described equations. As shown on Table 6, the value of R² (0.9979) corresponding to the pseudo-second-order was higher than that of pseudo-first order (0.9758). Therefore, the adsorption process follows the pseudo-second order which determines the nature of adsorption. The chemical interaction occurred to get rid of the cadmium ions, which confirmed the interpretation obtained previously; that at assortment of pH (6–8), an electrostatic interaction take place between the available cellulose anions and free cations in the aqueous solution.

Similarly, adsorption isotherms were modeled using the models of Dubinin–Radushkevich, Freundlich, Langmuir and Temkin. The obtained parameters were illustrated in Table 7. According to these values of R² and χ², the Langmuir seems as the best model to describe the adsorption process since its R² is equal to 0.9937, which was higher than that of the other three isotherms, Freundlich, Dubinin–Radushkevich and Temkin (Fig. 9). Langmuir isotherm describes more appropriately the adsorption process, which consists to a monolayer adsorption of cadmium onto cellulose (Foo and Hameed, 2010). Cellulose as a polymer with identical monomers chain, allowed the retention of Cd²⁺ on the same sites. This actively demonstrates that the cadmium ions adsorbs at sites identical of cellulose with same energy through formation of a molecular monolayer (Langmuir, 1916). The adsorption energy, E, was obtained from the relationship in below (Araissi et al., 2020; Faghihian et al., 2013):

Close

Fig. 9

Langmuir, Freundlich, Temkin and Dubinin-Radushkevich isotherms fitted to experimental sorption of Cd²⁺ onto cellulose.

Export to PPT

The obtained E value for adsorption of Cd²⁺ was 6.65 kJ/mol < 8.0 kJ/mol, indicating that the sorption process is done through physical sorption (Araissi et al., 2016; 2020; Faghihian et al., 2013).