3.1. Surface characteristics
The FTIR spectrum in Fig.1 (i) was analyzed in the 400–4000 cm-1 range to determine the interaction between each component during the CNTs/ZnO@laccase and CS-CNTs/ZnO@laccase synthesized process. As illustrated in Fig. (i, a), the formulation containing zinc oxide exhibit an important peak at 423 cm-1 assigned to the Zn-O band. In the (b) spectrum, a broad absorption band between 3200 and 3500 cm-1 is associated with hydroxyl functional groups on the surface of CNTs. The band around 521 cm−1 the FTIR spectrum of (c) indicates the presence of ZnO, indicating the successful formation of CNTs/ZnO nanocomposite [41].
The typical infrared spectrum of CNTs/ZnO@laccase is shown in the (c) spectra. Five leading bands detected the presence of the laccase-specific peaks [34]: (i) broad peaks near 3250 cm-1 indicated the interaction of -OH and -NH; (ii) weak bands near 2850 cm-1 attributed to C-H bonds; (iii) peaks at 1630 cm-1 attributed to C-O stretching vibrations; (iv) a band peaking at 1350 cm-1 assigned to amines′ C-N stretching vibrations; and (v) a strong and sharp band peaking at 1100 cm-1 ascribed to C-O-C groups. Additionally, the absorption peak at 3700 cm-1 was caused by the presence of (non-H-bonded) OH groups. After combining CNTs/ZnO@laccase and chitosan, the broader band around 3450 cm −1 was observed in Fig. (i, e), which is ascribed to the -OH and -NH groups in the chitosan chain (Fig. i, d). The peaks around 1650 cm−1 were related to the -NH scissoring from the primary amine ascribed to the cross-linked chitosan′s free amino groups. The characteristic peak at 1680 cm−1 was attributed to -NH2 stretching vibration, whereas the typical peak at 1431 cm−1 was associated with -CH bending vibration [42]. Another peak at 1146 cm−1 was associated with the anti-symmetrical stretching of the C-O-C bridge and vibration mode -CN stretching, while the peak at 1080 cm-1 was attributed with skeletal vibration involving C-O stretching, which are characteristics of a chitosan saccharide structure [43].
Fig. 1 (ii) depicts the XRD patterns of ZnO, CNTs, CNTs/ZnO@laccase, CS and CS-CNTs/ZnO@laccase. The conclusion can be drawn that the characteristic peaks of ZnO (32.08°, 34.5°, 36.3°,38.56°, 47.78°, 56.80° and 63.07°) corresponded to the crystal planes (100), (002), (101), (102), (110), (103), and (112) of the hexagonal wurtzite structure of ZnO NPs (JCPDS No. 36-1451), respectively. Additionally, the characteristic diffraction peak of CNTs was observed at 2θ values of 26°. Furthermore, the characteristic peak (10° to 30°) indicates the presence of a microstructure chitosan network. In the case of CS-CNTs/ZnO@laccase, strong and sharp characteristic peaks appeared, and all peaks associated with the CS crystalline regions disappeared, indicating a strong interaction between chitosan and CNTs/ZnO@laccase NPs.
FE SEM, and TEM were used to characterized the surface morphologies of the samples following coating with ZnO nanocomposite and laccase immobilization on the CNTs and when combined with chitosan. Fig. 2 (a, b) illustrated the preparation of ZnO nanoparticles and laccase on CNTs, respectively. The morphology of CNTs/ZnO@laccase coated with CS is depicted in Fig. 2 (c, d). Additionally, dot mapping results (Fig. 2 (e)) revealed a uniform dispersion of ZnO and laccase on the surface of CNTs and CS. Moreover, the selected area electron diffraction pattern (inset of Fig. 2 (f, g)) clearly shows ZnO nanoparticles and laccase on the surface of CNTs. Similarly, the EDS spectrum revealed that the prepared CS- CNTs/ZnO @laccase contains only 20% N, 32% O, 56% C, confirming the presence of elemental Zn and Cu signals associated with ZnO and laccase on the surface of CNTs.
Weak Zn and Cu signals, in comparision to C and O elements, may indicate lower Cu and ZnO values in the composite. The detailed examination of pure CNTs, CNTs/ZnO and CNTs/ZnO@laccase using a corresponding high magnification TEM (Fig. 2 (h), (i), and (j)) demonstrates that the CNTs are decorated with ZnO nanoparticles and laccase along their length.
3.2. Immobilization process
By subtracting the initial and final laccase activities divided by the initial amount in the reaction medium, the immobilization performance of laccase on CNT/ZnO nanoparticles was calculated. Laccase immobilization yield was approximately 83% in all experiments when laccase absorbed during the washing step was calculated. Each experiment was repeated three times with the average value used to obtain the same results. That is, laccase activity was 40 ± 1 U per 1mg of CNTs/ZnO nanoparticles. As a result, hydrogen bonds between the carboxyl groups of the CNTs/ZnO nanoparticles and the enzyme are expected to have been formed, resulting in laccase immobilization [44]. Afterward, the activity of 1mg immobilized laccase on CNTs/ZnO nanoparticles was determined directly, and yielding 43 ± 1.5 U, which was highly correlated with the amount mentioned previously. Indeed, another significant factor contributing to the enzyme′s activity is the electron transfer between laccase, nanotubes, and zinc oxide [36].
3.3. Thermal and storage stability of free and immobilized laccase
The study results on the storage and thermal stability of laccase, both free and immobilized, are shown in Figure 3. After ten days of storage, the immobilized laccase retained approximately 65% of its initial activity. In contrast, the free laccase was nearly inactive (Fig. 3(a)), indicating that immobilization may prevent enzyme denaturation and autodigestion that immobilized laccase on the nanocomposite support can act as a stable biocatalyst. The heat inactivation curves for both free and immobilized laccases demonstrated in Fig 3 (b), that the enzyme was more active at 20 °C than at 40 °C. Additionally, the difference in activity between free and immobilized laccase was insignificant at 20 °C. At 40 °C, however, there was a significant difference. Additionally, at 40 °C, immobilized laccase retained nearly 70% of its activity after 2h, compared to 20% for free laccase. These finding suggest the presence of a covalent bond between immobilized laccase and CNTs, as well as the presence of chitosan coatings that protect enzymes from denaturation and structural changes caused by heat. At elevated temperature, the nanocomposite support effectively protect laccase from inactivation [36,45].
3.4. Reusability of the immobilized laccase
Five days of cyclic tests were conducted to determine the reusability of the immobilized laccase and the CS-CNTs/ZnO @laccase nanocomposite. The percentage of BPA removed was calculated using the relative activity of immobilized laccase and the nanocomposites′ stability. After each series of experiments, the nanosorbent was washed and dried. As illustrated in Fig. 4, laccase′s relative activity decreased by 33% from the first to the fifth day, followed by a slight decrease in catalyst stability. After five runs, a 33% reduction in catalyst efficiency demonstrated the nanocomposite′s exceptional chemical stability. This indicates that CS-CNTs/ZnO@laccase nanocomposites may be used as a robust catalyst in refining processes. The decreased efficiency could be attributed to laccase leaching from the CS-CNTs/ZnO nanocomposites or laccase inactivation [46].
3.5. Swelling Behavior
Because Chitosan-based adsorbents swell in aqueous media, a simple method for estimating swelling can be found in the gravimetric determination of water. Fig. 5 shows the swelling behavior of chitosan biopolymers at pH=7 and 30 °C. As depicted in Fig 5, the swelling increases gradually over time, reaching a constant level after 450 minutes [39]. Both CS-CNTs/ZnO and CS-CNTs/ZnO@laccase hydrogels demonstrated less swelling when compared to pure chitosan. The reason for the decrease in swelling in the compounds could be one of the following: i) The presence of a non-porous network structure attributed to CNTs, as demonstrated by FE SEM image analysis, and the presence of CNTs in the polymer matrix reduces the polymer matrix′s hydrophilicity. ii) Because the chitosan matrix is destroyed and the hydroxyl groups are limited, the interface between chitosan and doped zinc oxide in CNTs disrupts the intermolecular attractiveness. iii) The swelling ability CNTs/ZnO@laccase water confirmed its amphiphilic nature. Amphiphilic CNTs/ZnO@laccase hydrogels improved mechanical properties due to the presence of hydrophobic fields, which decreased their swilling in water, and their structure became more cohesive [47].
3.6. The effect of adsorption
The effect of adsorption on BPA removal from aqueous was investigated using CS, CS-CNTs/ZnO nanoparticles, and CS-CNTS/ZnO@laccase Under the same conditions. As shown in Fig. 6, the adsorption percentages for various adsorbents are as follows: CS-CNTs/ZnO@laccase (91.53) ˃ CS-CNTs/ZnO nanoparticles (75.21) ˃˃ CS (30.72). It was demonstrated that CS-CNTs/ZnO@laccase nanocomposites are effective at removing BPA. The reason for this could be the large surface area and volume of the CS and CNTs structures, the increased bubble nucleation required for the cavitation process due to the presence of ZnO particles and CNTs, and the high activity of the immobilized laccase in the CS-CNTs/ZnO nanoparticles. In Continuation, the RSM method was used to investigated the removal of BPA by CS-CNTS/ZnO@laccase.
3.7. Modeling and optimization of BPA degradation using RSM
According to Section 2.9, the experimental results for 30 RSM-CCD-designed experiments are summarized in Table 3. The software′s predicted results are presented, and a quadratic polynomial model was estimated as a function of operational parameters (independent variables) and deletion percentage (dependent variable) and coded as Eq. ( 4 ). Analysis of variance was used to ensure the proposed model′s accuracy. The analysis of variance for the adsorption process by the nano biocatalyst obtained from the RSM model is shown in Table 3. The F-value of the model is 13.73, and the p-value is less than 0.0001, indicating that the proposed model is highly adequate. Additionally, the correlation coefficient (R2) of 0.9276 is consistent with the set value of 0.9015; for R2 (i.e., the difference is less than 0.05). There is a strong correlation between experimental and theoretical performance values. P-values less than 0.0500 indicate that model terms are significant and vice versa. The model terms A, B, C, D, BC, CD, A2, B2 and D2 are significant in this design. Eq. 4 was simplified to based on ANOVA results (Table 3) and the omission of nominal terms.
R1 = +80.32500 - 4.58125 [BPA]0 +7.47708 [Catalyst]0 + 5.6479 pH + 7.74958 Time -1.05062 [BPA]0 ⨯ [Catalyst] 0 - 0.59312 [BPA]0 pH - 0.9181 [BPA]0 Time - 5.24187 [Catalyst]0 pH - 0.93437[Catalyst]0 Time + 3.87063 pH Time - 2.96677 [BPA]02 - 4.03552 [Catalyst]02 - 1.59927 pH2 - 2.92677 Time2 Eq. (4)
Table 3. Analysis of variance for the adsorption process derived from the RSM model
Source
|
Sum of Squares
|
df
|
Mean Square
|
F Value
|
p-value Prob > F
|
Model
|
5515.94
|
14
|
394.00
|
13.73
|
<0.0001
|
A-[BPA]0
|
503.71
|
1
|
503.71
|
17.55
|
0.0008
|
B-[Catalyst]0
|
1341.76
|
1
|
1341.76
|
46.75
|
<0.0001
|
C-pH
|
765.58
|
1
|
765.58
|
26.67
|
0.0001
|
D-Time
|
1441.35
|
1
|
1441.35
|
50.22
|
< 0.0001
|
AB
|
17.66
|
1
|
17.66
|
0.62
|
0.4450
|
AC
|
5.63
|
1
|
5.63
|
0.20
|
0.6642
|
AD
|
13.49
|
1
|
13.49
|
0.47
|
0.5035
|
BC
|
439.64
|
1
|
439.64
|
15.32
|
0.0014
|
BD
|
13.97
|
1
|
13.97
|
0.49
|
0.4961
|
CD
|
239.71
|
1
|
239.71
|
8.35
|
0.0112
|
A2
|
241.42
|
1
|
241.42
|
8.41
|
0.0110
|
B2
|
446.69
|
1
|
446.69
|
15.56
|
0.0013
|
C2
|
70.15
|
1
|
70.15
|
2.44
|
0.1388
|
D2
|
234.95
|
1
|
503.71
|
8.19
|
0.0119
|
Residual
|
430.55
|
15
|
28.70
|
|
|
Lack of Fit
|
320.61
|
10
|
32.06
|
1.46
|
0.3549
not significant
|
Pure Error
|
109.94
|
5
|
21.99
|
|
|
Cor Total
|
5946.49
|
29
|
|
|
|
Quality of quadratic model
|
R2
|
0.9276
|
Adjusted R2
|
0.9015
|
When the residuals are normal with zero mean, constant variance, and independent variance, the proposed model is confirmed to be adequate. Fig. 7 (a) depicted the normal plot of residual versus internal residual, while Fig. 7 (b) illustrates the plot of residual versus runs. The first figure demonstrated that the error variance was uniform and that there was no apparent scattering. The second graph established the validity of the chosen model for examining the relationship between the obtained response and the independent variables [48].
Following statistical analysis and validation of the proposed model′s accuracy, three-dimensional graphs (response levels) were used to evaluate the effect of independent variables ([BPA], catalyst amount, pH, and irradiation time) on the response rate (percentage of removal), depicted in Fig. 8. After examining the role of operational variables in removing the highest BPA concentration from water, using an aromatic hydrocarbon model, DX7 software was used to performed an optimization process. The theoretical efficiency of BPA removal was 89.86% under optimal conditions (initial BPA concentration of 100 mg L-1, catalyst concentration of 1 g L-1, pH= 9, and 40 min of contact exposure) (run 15 in Table 2). Experiments conducted in optimal conditions revealed that approximately 94.51% of BPA was destroyed, confirming the accuracy and validity of RMS′s proposed model.
According to studies, decreasing the amount of BPA and increasing the irradiation time and catalyst concentration, increases the number of electrons and holes produced, resulting in the formation of more active species (OH°, O2°-) and increasing the degradation efficiency. Superoxide radicals (O2°-), hydroxide radicals (OH°), and produced holes (h+) are the major species involved in the photocatalytic degradation of BPA. Based on the literature, OH° and h+ were more effective at decomposition than O2°- when combined with CS-CNTs/ZnO @laccase nanocomposites. Numerous studies have been conducted on the effect of pH on the electrolytic degradation of aromatic organic compounds. According to some researchers alkaline and acidic conditions are optimal for decomposing aromatic organic compounds, while neutral conditions are ideal [49,50].