FTIR analysis
The FTIR spectra of BS, BBS, NC, PZ, FZ1, and CFZ1-2 were scanned in the range of 4000 – 400 cm−1 and the recorded spectra are illustrated in Fig. 2. The broadbands observed at 3408 cm−1 in all samples belong to the stretching vibration of O-H and the peak at 1646 cm−1 is attributed to the absorbed water molecules by the samples (Langari et al. 2019). These peaks show the hydrophilic tendency of all samples. As shown in Fig. 3(a), the band located around 2894 cm−1 is assigned to the C-H stretching vibration in cellulose structure (Lefatshe et al. 2017). The bands at 896 and around 1060 cm−1 in all spectra are attributed to the C-H rocking and C-O stretching vibrations of cellulose, respectively (Chandra et al. 2016). In addition, the peak at 1160 cm−1 is related to the stretching of C-O-C in the β-1,4-glycosidic linkages between D-glucose units in cellulose (Lee et al. 2017). The increase in the intensity of these peaks from BS to isolated NC shows the enhancement of the cellulose content. The peak observed at 1732 cm−1 in the spectrum of BS is assigned to the presence of C=O stretching vibration of acetyl and carbonyl groups in hemicellulose or the ester linkage of carboxylic group in the ferulic and p-coumaric acids of lignin and/or hemicellulose (Lee et al. 2017). Additionally, the peaks at 1508 and 1254 cm−1 are associated with the C=C and C-O stretching in aromatic skeletal vibrations of lignin, respectively. These peaks disappeared in the spectra of BBS and NC, indicating that the lignin content was decreased during bleaching and NC isolation (Chandra et al. 2016). The band at1270 cm−1 in the spectrum of BS is referred as the C-O stretching vibration of hemicellulose (Shahbazi et al. 2017). However, the reduction of this peak in the spectrum of BBS indicates the hemicellulose removal after bleaching. These results indicate that alkali and bleaching treatments were effective to extract highly purified cellulose. Moreover, the increase in the intensity of characteristic peaks of cellulose shows that the structure of cellulose was not degraded after alkali pretreatments, bleaching and acid hydrolysis (Shahbazi et al. 2017). The FTIR spectra of NC, PZ, FZ1, and CFZ1-2 are presented in Fig. 3(b). The characteristic peaks around 430-500 cm−1 correspond to metal-oxygen vibration, which are observed in the spectra of PZ, FZ1 and CFZ1-2. It is worth noting that the characteristic peaks of FZ1 are similar to those of PZ, indicating that the doped Fe into ZnO existed in the form of elements. Similar characteristic peaks of NC were observed in the spectrum of CFZ1-2 composite. Moreover, compared with NC, a new absorption peak at 400–500 cm−1 assigned to Zn-O or Fe -O stretching vibration was found in the CFZ1-2 composite (Guan et al. 2019). Furthermore, it could be seen that the intensity of these characteristic peaks decreased in the spectrum of CFZ1-2. This reduction confirms the strong interaction between FZ particles and NC chains.
Morphology and dimensions
Figure 3. displays the FESEM and TEM images of as-prepared samples. As shown in Figs. 3(a1-a3), before alkali pretreatment and bleaching, the BS showed a rigid uniform structure and the fibers of the BS were fully cemented with hemicellulose and lignin. Importantly, the surface of samples was rough, probably due to the presence of lignin. However, after alkali treatment and bleaching, the BS fibers exhibited a loose structure due to the degradation of lignin and hemicellulose removal. After ultrasonic-assisted acid hydrolysis, the obtained NC revealed nearly spherical nanostructure with an average particle diameter of 70 nm (Figure 3(a3)).
Figures 3(b1-b4) show the surface morphologies of the pristine and Fe-doped ZnO. As shown in the FESEM images, the average size of the particles was in the nanometer range. It is evident from Fig. 3(b1) that PZ sample presented a nearly zero-dimensional (0D) growth with the diameter of 100 nm and Fe3+ doping affected the growth direction and the shape of the synthesized nanoparticles. In the FZ1, FZ3, and FZ5 samples, the 0D growth of PZ gradually turned into 1D growth and the particle length increased to 120±10, 140±10, and 150±10 nm, respectively. The particle diameter was slightly influenced by the dopant concentration. As the dopant concentration increased from 1 to 5%, the diameter decreased from 50±10 to 30±10 nm. The surface morphology of CFZ hybrid nanocomposites is shown in Figures 3(c1-c4). As can be seen, by the addition of NC during the synthesis of FZ1, the average diameter of nanocomposite increased to 70±10, 75±10, 85±10, and 100±10 nm for CFZ1-1, CFZ1-2, CFZ1-3, and CFZ1-4 samples, respectively. Besides, due to the greater tendency of NC to form a precipitate, with increasing its content in the nanocomposite, larger sediments were observed in the nanocomposites.
Figures 3(d1-d4) display the TEM images of NC, PZ, FZ1, and CFZ1-2, respectively. The image of the NC sample clearly revealed that the particle size distribution of NC was in the range of 20 to 150 nm which were scattered next to each other (Fig. 3(d1)). The TEM image of PZ sample evidently exhibited the spherical morphology of the nanoparticles (Fig. 3(d2)). It is clear that Fe3+ doping changed the morphology of the PZ particles (Fig. 3(d3)). Further, the TEM image of CFZ1-2 composite illustrated in Fig. 3d4 identifies that particles were fabricated with successful integration between Fe-ZnO and NC.
XRD analysis
The XRD analysis was carried out to identify the crystalline structure of samples. The diffraction patterns of the BS and NC samples are shown in Fig. 4a. The patterns of both samples exhibited typical cellulose I peak around 2θ = 16.1°, 22.5°, and 34.1° attributed to (110), (200), and (004) planes, respectively (Abdalkarim et al. 2018; French 2014). The crystallinity index (CI) is an important parameter that influences the physical and mechanical properties of cellulose. From the XRD profiles, the CI of BS and NC was calculated using the Segal equation as follows (Gedik 2021):
\(CI\text{\%}=\frac{{I}_{mi}-{I}_{am}}{{I}_{mi}}\times 100\)
|
(4)
|
where Imi is the maximum intensity of the principal peak (200) and Iam is the value of the intensity minimum of the peak (110) at 2θ = 16.1°, which represents the amorphous fraction. The calculated CI was 40.27 and 60.16% for the BS and NC, respectively. According to the results, it was found that the treatment steps affected the crystallinity of the BS by the removal of excess amorphous regions, such as lignin, hemicellulose, and the amorphous region of cellulose from the BS fibers.
The XRD patterns of PZ and FZs are shown in Figs. 4(b-c). As shown in the XRD patterns, PZ and FZs samples displayed many intense peaks at 2θ = 31.78°, 34.41°, 36.25°, 47.53°, 56.59°, 62.83°, 66.39°, 67.64°, 69.05°, 72.53°, and 76.92° which assigned to the (100), (002), (101), (102), (110), (103), (200), (112), (201), (004), and (202) planes of the hexagonal wurtzite structure, respectively, and resemble those previously reported (Isai and Shrivastava 2019). On the basis of wurtzite structure, the PZ and FZs samples correspond to the standard data (JCPDS No.01-080-0074). Meanwhile, no additional peaks were detected in FZs patterns, which indicated that iron was well doped into the crystal phase of PZ. The XRD diffractograms of CFZs nanocomposite are illustrated in Fig. 4(d). The CFZs patterns exhibited two-phase structures corresponding to FZ and NC. It can be deduced that the presence of NC did not alter the crystal structure of FZ, and Fe-doped ZnO was crystallized well in the NC polymeric matrix, as no new peaks were detected.
The average crystallite size (D) of the PZ, FZs and CFZs was calculated using the Scherrer formula as:
\(D=\frac{0.89\lambda }{\beta \text{cos}\theta }\)
|
(5)
|
where \(\lambda\)is the wavelength of the X-ray radiation (0.15406 nm), β is the full width at half maximum (FWHM) measured in radians, and \(\theta\) is the Bragg diffraction angle.
The (110) peak, the most intense peak for PZ, FZs, and CFZs samples, was used for the calculation of the crystallite sizes. The calculated crystallite size for PZ, FZs, and CFZs is summarized in Table 2. The average crystallite size of PZ was 17.8 nm. Results showed that the Fe-doped ZnO nanoparticles with higher content of Fe revealed smaller average crystallite sizes. Results showed that the substitution of Fe3+ (0.64 Å) in place of Zn2+ (0.74 Å) in zinc oxide lattice reduced the growth rate of zinc oxide crystallites as observed in other studies (Ciciliati et al. 2015; Saleh et al. 2012; Xiao, Zhang, Wei, and Chen 2018). The average crystallite size of CFZ1-1, CFZ1-2, CFZ1-3, and CFZ1-4 was 12.7, 11.0, 9.1, and 8.2 nm, respectively. Compared to the crystallite size of FZ1, the crystallite size of CFZs reduced which is due to the inhibition of NC from the nucleation and growth of ZnO nanocrystals (Fabbiyola et al. 2016). The NC as a matrix could disperse Fe-doped ZnO particles and further decrease the crystallite size of Fe-doped ZnO.
Table 3
The calculated average crystallite size of PZ, FZs, and CFZs
Samples
|
Composition
|
Average crystallite size (nm)
|
PZ
|
Pure ZnO
|
17.8
|
FZ1
|
1% Fe doped ZnO
|
15.5
|
FZ3
|
3% Fe doped ZnO
|
14.4
|
FZ5
|
5% Fe doped ZnO
|
13.6
|
CFZ1-1
|
1% Fe doped + 0.1 NC blended ZnO
|
12.7
|
CFZ1-2
|
1% Fe doped + 0.2 NC blended ZnO
|
11.0
|
CFZ1-3
|
1% Fe doped + 0.3 NC blended ZnO
|
9.1
|
CFZ1-4
|
1% Fe doped + 0.4 NC blended ZnO
|
8.2
|
UV-vis DRS analysis and bandgap study
The influence of Fe3+ doping on optical properties of ZnO was studied using UV–vis DRS analysis. The diffuse reflectance spectra of the pure ZnO and FZs are displayed in Fig. 5(a). It was observed that the pristine ZnO nanoparticles exhibited a low diffuse reflectance at wavelengths less than 400 nm, which corresponds to an intense absorption peak between 200 to 400 nm. This result indicates that the PZ can efficiently absorb UV light. Compared to the pure ZnO, the FZs samples had lower diffuse reflectance in the visible region. On the other hand, Fe-doped samples were able to efficiently absorb visible light (solar light) for the photocatalytic process. Additionally, the bandgap energy of all samples was calculated from the well-known Kubelka-Munk (KM) equation and the plots are illustrated in Fig. 5(b). The calculated bandgap energy of the FZ1, FZ3 and FZ5 samples are approximately 3.09, 3.03, and 3.01 eV, respectively, which is smaller than that of PZ (3.22 eV). This reduction could be attributed to the s-d and p-d exchange interactions between the band electrons and the d electrons of the Fe3+ ions substituting Zn2+ ions (Ciciliati et al. 2015).
Thermal stability
The TGA and DTG curves of NC, PZ, FZ1, and CFZ1-2 samples are depicted in Fig. 6. The PZ sample exhibited approximately 2.0% initial weight loss below 100°C, which corresponds to the evaporation of the adsorbed water on the surface of PZ nanoparticles. Then, subsequent weight loss of 4.3% occurred at 260-320°C, which was attributed to the removal of organic components. The FZ1 sample displayed the highest thermally stability. A total weight loss of 2.6% was found for FZ1 sample. As observed, the thermal degradation of NC and CFZ1-2 samples took place in two stages. The first weight loss of 5% for NC sample occurred at about 250°C; this was followed by a second weight loss of 60%, which happened at around 330°C. The thermal weight loss of NC during this stage was mainly attributed to the depolymerization, dehydration, and decomposition of glycosyl units of NC sample. Compared with the NC sample, the thermal degradation peaks for CFZ1-2 were found at higher temperatures (290 and 350°C). This was related to the stronger interactions between oxygen atoms of the NC and FZ1 nanoparticles that provide a thermal barrier for the NC skeleton. The TGA analysis showed a total weight loss of 22.76% for CFZ1-2. These results indicate that the thermal stability of CFZ1-2 was better than that of NC.
Photocatalytic degradation of MB
The photocatalytic activity of PZ and FZs was investigated in the MB degradation reaction, when exposed to both UV and visible light irradiations. In these experiments, the effect of iron ions in the photocatalysts on the degradation of MB was evaluated. In order to minimize the influence of adsorption of the synthesized catalyst, each experiment was performed in the dark condition for 30 min prior to the light irradiation. Fig. 7(a) shows the photocatalytic degradation of MB in an aqueous solution with PZ and FZs nanoparticles under UV light radiation. It can be observed that the PZ sample degraded 93.13% of the MB dye in 150 min under UV light radiation, but the prepared Fe-doped samples exhibited lower photocatalytic activity. As the dopant molar ratio increased from 1 to 3 and 5%, the degradation of MB correspondingly decreased from 86.34 to 73.81 and 66.25%, respectively. As shown in Fig. 7(b), under UV light irradiation, the Fe ions in the ZnO crystal structure act as an electron-hole trap and promote the charge recombination. Thus, increasing the Fe dopant ratio reduces the photocatalyst activity by disrupting the redox process.
Figure 8(a) illustrates the photocatalytic degradation of MB in an aqueous solution with PZ and FZs nanoparticles under visible light radiation. The degradation efficiency of MB over the PZ catalyst was 39.96% in 150 min of visible light irradiation. The pristine ZnO revealed poor photocatalytic performance under visible light irradiation compared to that under UV light irradiation. This could be attributed to the high bandgap of ZnO. As observed in Figure 8(a), the FZ1 catalyst exhibited the highest degradation efficiency of MB (94.21%), followed by FZ2 (83.97%) and FZ3 (50.62%). As expected, the increase of Fe3+ doping ratio was very effective to enhance the performance of ZnO. A number of researches confirmed that the photocatalytic degradation efficiency of ZnO can be enhanced by doping with iron ions (Xiao, Zhang, Wei, Yu, et al. 2018; Yi et al. 2014). A scheme of plausible mechanism for the MB photocatalytic degradation over Fe-doped ZnO under visible light is displayed in Figure 8(b). According to the previous studies (Yi et al. 2014), the photodegradation of dye under visible light irradiation is not due to the direct reaction between photogenerated electron holes on the surface of catalyst and dye, but can be attributed to the photosensitization effect of the dye molecules. During this process, the MB molecules could absorb visible light in the range 450-600 nm and get excited by the visible light. Then, the electrons on the highest occupied molecular orbital (HOMO) of MB molecules transfer to the lowest unoccupied molecular orbital (LUMO) and are immediately injected to the conduction band (CB) of ZnO, whereas MB is converted to its radicals (MB•+) form. When iron ions are added into the ZnO structure, they could increase the electron capture capacity of ZnO from the photosensitizer (MB). This process could absorb more photons and effectively enhance the separation of the electron–hole pairs. The transferred electrons to the CB could react with electron acceptors such as the adsorbed O2 on the surface of the catalyst to make superoxide radical anion O2•−, which are further converted to HO•, H2O2 and HO• species via a series of reactions. However, when the doping ratio was 3 and 5%, the formed ZnFe2O4 occupies the active sites and hinders the electrons transfer to Fe3+; thus, the electron–hole pairs could not be separated effectively, so the photocatalytic activity is gradually reduced to 83.97 and 50.62% for FZ3 and FZ5, respectively (Türkyılmaz et al. 2017; Yi et al. 2014).
The prepared CFZ samples were also utilized to evaluate their dye removal performance; the results of this investigation are exhibited in Fig. 9(a). Results showed that the dye removal at the dark condition was influenced by the NC content in nanocomposite. The dye removal increased with an increase in NC /Zn2+ molar ratio from 0.1 to 0.4. The possible mechanism of MB degradation during this time is physical adsorption that relies on the surface hydroxyl or sulfate ester groups of NC and electrostatic attraction between NC and MB (An et al. 2020). MB is a positively charged cationic dye, while the surface charge of CFZs was negative; this could form a strong electrostatic interaction between them. The dye removal efficiency of CFZ1-1, CFZ1-2, CFZ1-3, and CFZ1-4 at the end of the dark time was 25.53, 33.99, 40.09 and 51.33%, respectively. Compared to other nanocomposites, CFZ1-4 with a higher molar ratio of NC to Zn+2 showed the highest dye removal efficiency. The phototocatayltic activity of CFZs was evaluated for MB degradation under visible light. It can be observed that the CFZ samples possessed photocatalytic ability to degrade the MB under visible light irradiation. Based on the obtained results, the performance of CFZ1-1 with a degradation efficiency of 86.61% was weaker than that of CFZ1-2 with 97.81% degradation efficiency, which was probably related to the incomplete incorporation of NC in photocatalyst particles in the CFZ1-1 sample. With further increase in the molar ratio of NC to Zn+2 from 0.2 to 0.4, there was a reduction in the degradation efficiency. This might be due to the low absorption of light by NC and the coverage of photocatalytic particles by this biopolymer which limited the photocatalytic activity of these particles. The degradation efficiency of CFZ1-3 and CFZ1-4 was 69.34 and 54.34%, respectively, which was lower than that of CFZ1-2. According to the results, the multifunctional CFZ1-2 sample showed the best combination of adsorption and photocatalytic activity among the PZ, FZs, and CFZs samples. Therefore, CFZ1-2 was used to study its photocatalytic performance for the degradation of MB in the subsequent experiments.
Effect Of Solution Ph
As reported in several studies, adsorption facilitates the photocatalytic degradation process; higher adsorption results in a greater effectiveness of the photocatalysis process (Kumar et al. 2018; Tran et al. 2021). The dye molecules adsorbed on the catalyst surface are degraded faster than those in bulk due to the short life and easy annihilation of charge carriers or reactive radicals during diffusion or migration (Tran et al. 2021). By influencing the charging behavior of the catalyst and dye, the pH of the solution is an important factor that affects the dye photocatalytic degradation. The concentration of hydroxyl radicals, charge of the molecule, adsorption/desorption of the dye molecule and its intermediates onto a photocatalyst surface, and the surface charge property of the photocatalyst depend upon the pH of the dye solution. In order to determin the surface charge of CFZ1-2, the pH of zero-point charge (pHpzc) was determined (data not shown). It was obtained to be 8.1, which indicates that the surface charge of the catalyst is positive when pH<8.1 and vice versa. The solution pH was varied from 3 to 11 to study its effect on the photocatalytic degradation of the MB dye. Fig. 10 (a) indicates the effect of solution pH on the MB degradation. Results revealed that CFZ1-2 could act effectively over a wide range of pH (from 3 to 11). As expected, CFZ1-2 exhibited the highest degradation performance at pH 9, followed by pH 11, 7, 5, and 3. At acidic pH, the degradation of MB was unfavorable due to the strong electrostatic repulsion between CFZ1-2 and MB molecules. Besides, the acidic condition could increase the competitive adsorption of H+ with MB, hindering the MB access to the catalyst surface. The high degradation of MB may be resulted from the electrostatic attraction between the negative surface charge of CFZ1-2 and the positive charge of cationic dye (MB), while the pH of the solution was above the pHpzc (8.1). Thus, pH 9 was selected as the optimum pH for MB degradation.
Effect of nanocomposite dosage
The effect of catalyst dosage on MB degradation efficiency was studied with different amounts of CFZ1-2 (0.5 to 2.0 g/L) after 90 min irradiation under visible light. As shown in Fig. 10 (b), the degradation efficiency enhanced with raising the catalyst dosage up to 1.5 g/L and reduced thereafter. An explanation for the observed trend could be attributed to the number of active sites on the catalyst, which enhanced with the addition of the photocatalyst, improving the degradation efficiency. However, an increase of the catalyst dosage to over 1.5 g/L had negative effect on the catalyst performance and led to the reduction of the degradation efficiency of CFZ1-2. It seems that the enhancement of catalyst amounts could lead to the agglomeration of photocatalyst particles, light scattering, and hindering the light transmission, hindering the photocatalytic performance (Emadian et al. 2020; Mirzaeifard et al. 2020). Based on the results, the optimum dosage of CFZ1-2 for MB degradation was selected at 1.5 g/L for the subsequent experiments.
Effect of initial dye concentration
The influence of initial dye concentration (5-25 ppm) on dye degradation by 1.5 g/L of the CFZ1-2 photocatalyst at pH= 9 was studied. As can be observed in Fig. 10(c), by increasing the dye concentration, the photocatalytic degradation indicated a downward trend. At a constant amount of catalyst, the number of active sites in the catalyst reduced upon dye adsorption onto the surface of catalyst, resulting in a decrease in the active radical production and also reduction of degradation of efficiency. Furthermore, when the MB concentration was increased, the turbidity of reaction mixture also increased that acted as a screen against visible light, prevented the light photons from reaching the active sites of the catalyst and deteriorated the photocatalytic degradation (Raja et al. 2020).
Regeneration
The reusability and stability of a photocatalyst are critical parameters for its practical application. The reusability of the CFZ1-2 as optimal photocatalyst was investigated under optimized conditions for the MB degradation (pH=9, catalyst dosage= 1.5 g/L, and MB initial concentration=5 ppm). The reusability tests were done by centrifuging the CFZ1-2 nanocomposite sample from the first cycle, which was then washed and oven-dried at 50°C. Fig. 11 illustrates that the degradation efficiency of the CFZ1-2 sample slightly decreased from 98.84 to 97.51, 95.39, 94.11, and 92.31% during the different reuse runs. In fact, the degradation efficiency of CFZ1-2 was kept higher than 92% after five successive runs. This negligible reduction of degradation efficiency might be attributed to the trapping of MB molecules on the active sites of the catalyst surface and the loss of the catalyst throughout the cycling experiments. The obtained result confirms that the synthesized photocatalyst is a promising catalyst for long-term uses in industrial wastewater treatment processes.