3.1. Characterization of synthesized nanocomposites
The synthesized samples were characterized with several techniques as follows:
3.1.1- XRD patterns
Figure 1 (a–c) illustrates the XRD patterns of Co/Ni/Al-LTH, ZIF-8 and [email protected] nanoparticles respectively. In Fig. 1a, the sharp diffraction peaks positioned at 2θ = 11.80◦, 19.50◦, 23.20◦, 38.95◦, 52.40◦, 59.60◦, 61.30◦ and 62.60◦ associated with the Miller indices of (003), (006), (009), (018), (110), (113), (012) and (015) respectively represent the crystallographic structure of Co/Ni/Al-LTH. The peaks at 11.80◦, 19.50◦ and 23.20◦ ascribed to Miller indices of (003), (006), and (009) indicate the stacking order of layers in LTH [26].
Figure 1b displays the XRD pattern of ZIF-8 peaks located at 2θ = 6.55◦, 10.47◦, 12.60◦, 14.50◦, 16.30◦, 17.90◦, 22.05◦, 24.45◦, 25.55◦, 26.50◦, 27.65◦, 28.90◦, 29.55◦, 30.55◦, 31.35◦, 32.25◦, 34.85◦ and 36.35◦ are associated with the Miller indices of (011), (002), (112), (022), (013), (222), (114), (233) and (224), respectively. By considering and comparison this result with the previous reports [27, 28] it can be said, this compound was well synthesized. Figure 1c exhibits the XRD pattern of the Co/Ni/Al-LTH @ZIF-8 nanocomposite, as can be observed this pattern almost consists of both peaks in XRD patterns of Co/Ni/Al-LTH and ZIF-8. Results reveal that the combination of Co/Ni/Al-LTH and ZIF-8 nanomaterials does not disturb their regular structure and the nanocomposite was successfully synthesized.
3.1.2. Studies of FT-IR
Figure 2(a-c) illustrates FT-IR spectra of Co/Ni/Al-LTH, ZIF-8 and Co/Ni/Al-LTH/ZIF-8 composite. Figure 2a shows the FT-IR spectrum of the Co/Ni/Al-LTH. The narrow peak at 3636 cm− 1 can be attributed to the non-hydrogen-bonded hydroxyl groups. the broad peak around 3568 cm− 1 can be assigned to the O–H stretching modes of the interlayer water molecules and the H-bound O–H groups and the peak at 1714 cm− 1 is attributed to the bending mode vibration of the interlayer water molecule. The absorption around 1419 cm− 1 can be assigned to the intercalated carbonate anions [29]. Absorption bands observed in 1093, 1223 and 1363 cm− 1 can be ascribed to interlayer nitrate groups [30]. Absorption bands around 1000 cm− 1 are due to the stretching and bending vibrations modes of M-O and M-OH. (M = Co2+, Ni2+, Al3+)
Figure 2b shows the FT-IR spectrum of the ZIF-8 metal organic network. The bands observed at 689 and 755 cm − 1 can be ascribed to the stretching vibrations modes of the Zn-N and Zn-O bonds, respectively [31]. Also, the bands related to 1143 ,1584 and 2927 cm − 1 can be associated to the C-N, C = N and C-H stretching vibrations modes in the imidazole ring, respectively. Figure 2c exhibits the FT-IR spectrum of the Co/Ni/[email protected] nanocomposite as can be observed this spectrum almost includes of both signals in spectra of (2a) and (2b).
3.1.3. Morphology studies of the synthesized samples
The surface morphology of the Co/Ni/[email protected] was examined by field emission scanning electron microscopy. Figures 3a -3c illustrate FESEM images of the Co/Ni/Al-LTH, ZIF-8 and Co/Ni/Al-LTH @ ZIF-8, respectively and Figs. 3d and 3e show the EDX spectra of the ZIF-8 and Co/Ni/Al-LTH @ ZIF-8. Figure (a) shows the images of Co/Ni/Al-LTH nanosheets as hexagonal nanoplates with a size of about 100–200 nm. Figure 3 (b) shows images of ZIF-8 particles with a uniform polygonal structure measuring approximately 30–60 nm. Figure 3 (c) exhibits the FESEM images of the Co/Ni/Al-LTH @ ZIF-8 composite. The morphology of Co/Ni/Al-LTH @ ZIF-8, has a moderately change with the Co/Ni/Al-LTH images, transforming from hexagonal nanoplates to spherical shape (Fig. 5a & b). The EDX spectra with the tables of elemental composition of EDX confirm that ZIF-8 and Co/Ni/Al-LTH @ ZIF-8 composite are presented in Fig. 3 (d,e). Results of EDX confirm that the Co/Ni/Al-LTH @ ZIF-8 nanohybrid was successfully synthesized.
3.1.4. Thermal analysis
Results of thermal analysis of Co/Ni/Al-LTH, ZIF-8 and Co/Ni/Al-LTH/ZIF-8 are presented in Fig. 4(a-c). Figure 4(a) shows thermogram of Co/Ni/Al-LTH that contains two steps. In the first step, due to the evaporation of water molecules present on the surface and as well as interlayer water molecules, within the range of 50–280°C temperature. In the second step, weight loss observed at a temperature range of 280–350°C is ascribed the dehydroxylation of the brucite-like metal layers and the release of intercalated anions [32].
Figure 4(b) presents the thermogram of ZIF-8, a graduate weight loss owing to evaporation of water molecules absorbed on the surface and in cavities of the compound below 200°C. The second weight-loss stage belongs to decomposition of the organic ligand 2-methylimidazole. Eventually, degradation of the metal organic framework is the reason for the weight loss of the third stage [33]. Figure 4c shows thermal behavior of the Co/Ni/[email protected] composite. The weight loss of the first and second stages related to solvent extraction and dehydroxylation in Co/Ni/Al-LTH nanoparticles and the third stage is related to the decomposition of the ZIF-8 metal organic framework in the Co/Ni/[email protected] composite.
Figure 5 shows the surface charge evaluation of the adsorbents synthesized in this study through the zeta potential. The zero-point charge (zpc) of nanocomposit particles is around 10. This value is very close to the zpc of the ZIF-8 [34]. This pronounces that Co/Ni/Al-LTH is covered by ZIF-8 particles and reconfirming above results. The adsorbent surface will be negatively charged surfaces when pH > pzc and in contrast will be positively charged when pH < pzc. Hence the adsorption process is influenced by the repulsion or attraction forces between adsorbent and adsorbate, this phenomenon due to the pzc of the adsorbent (i.e., Co/Ni/Al-LTH /ZIF-8) and pKa of the adsorbate (i.e., diazinon) and also pH of the solution. As a result, to find higher removal efficiency, the pH value of the solution would be optimized.
3.1.5. Adsorption isotherm studies
Figure 6 (a-c) shows the Nitrogen adsorption/desorption (BET method) and pore size distribution curves (BJH method) of Co/Ni/Al-LTH, ZIF-8 nanoparticles and Co/Ni/[email protected] composite, respectively. According to the IUPAC classification, the synthesized Co/Ni/Al-LTH can be categorized as type IV with H4 hysteresis loop (Fig. 6a), which indicates the existence of slit-like mesoporous structures of the corresponding compound [37]. As shown in Fig. 6b, the ZIF-8 nanoparticle isotherm represents type I, which is characteristic of microporous materials. Figure 6c is similar to the Fig. 6b indicate that the nanohybride of Co/Ni/[email protected] consists microspore, and also its isotherm is classified as type I.
Table 1 presents the total surface area (SBET) and total pore volume for the synthesized LTH, ZIF-8 and nanocomposites, respectively. Both surface area and total pore volume of Co/Ni/[email protected] is very close to the ZIF-8, nevertheless both aforementioned parameters are significantly more than that for Co/Ni/Al-LTH.
Table 1
BET results of the synthesized samples.
Adsorbent
|
SBET (m2g− 1)
|
Total pore volume (cm3g− 1)
|
Mean pore diameter (nm)
|
Co/Ni/Al-LTH
|
28.8
|
0.027
|
3.8
|
ZIF-8
|
1370
|
0.523
|
1.52
|
Co/Ni/[email protected]
|
1030
|
0.401
|
1.54
|
3.2. Design expert and optimization
An experimental design method was planned using Design Expert 13 software in which the surface response design technique is employed based on the Box–Behnken method. The parameters of adsorbent dose, pH, and contact time were selected as the main affecting factors for removal of diazinon. Table 2 shows three levels (low (1−), medium (0), and high (+ 1)).. Table 3shows the Box-Behnken design matrix by considering three factor-three levels of input parameters and the corresponding output response for the removal of diazinon by the Co/Ni/Al-LTH /ZIF-8.
Table 2
Experimental range and level of factors in Box-Behenken Design.
Factors
|
Levels
|
|
(-1)
|
(0)
|
(+ 1)
|
Adsorbent dose (mg)
|
10
|
20
|
30
|
pH
|
4
|
7
|
10
|
Contact time (min)
|
10
|
20
|
30
|
Table 3
Box-Behenken design matrix for three factors along with observed predicted responses.
Run
|
Adsorbent dose (A)
|
pH (B)
|
Contact time (C)
|
Removal of DIZ (%)
|
1
|
10
|
10
|
20
|
60.76
|
2
|
20
|
4
|
10
|
64.27
|
3
|
10
|
7
|
30
|
71.29
|
4
|
20
|
7
|
20
|
82.69
|
5
|
20
|
7
|
20
|
82.61
|
6
|
30
|
4
|
20
|
68.65
|
7
|
20
|
7
|
20
|
81.71
|
8
|
20
|
10
|
10
|
73.04
|
9
|
20
|
7
|
20
|
82.69
|
10
|
10
|
7
|
10
|
71.29
|
11
|
30
|
10
|
20
|
69.53
|
12
|
20
|
7
|
20
|
81.82
|
13
|
30
|
7
|
10
|
83.57
|
14
|
10
|
4
|
20
|
53.74
|
15
|
20
|
10
|
30
|
67.8
|
16
|
20
|
4
|
30
|
67.67
|
17
|
30
|
7
|
30
|
83.98
|
In this study, results of the analysis of variance (ANOVA) were performed using Design Expert 13 software. In accordance with the conventional acceptance of statistical significance, at a 95% confidence level, the p-value must be less than 0.05 to be meaningful. If the p value for non-compliance is greater than the value selected for significance at a confidence level, it indicates that the model is desirable [35].
According to the p values obtained from the results of ANOVA (Table 4), all three factors considered in the removal of diazinon was significance.
Table 4
Result of analysis of variance (ANOVA)
Source
|
Sum of Squares
|
df
|
Mean Square
|
F-value
|
p-value
|
|
Model
|
1351.19
|
9
|
150.13
|
540.45
|
< 0.0001
|
significant
|
A
|
295.85
|
1
|
295.85
|
1065.02
|
< 0.0001
|
|
B
|
35.28
|
1
|
35.28
|
127.00
|
< 0.0001
|
|
C
|
0.2556
|
1
|
0.2556
|
0.9202
|
0.3694
|
|
AB
|
9.42
|
1
|
9.42
|
33.93
|
0.0006
|
|
AC
|
0.0420
|
1
|
0.0420
|
0.1513
|
0.7089
|
|
BC
|
18.66
|
1
|
18.66
|
67.18
|
< 0.0001
|
|
A²
|
101.02
|
1
|
101.02
|
363.66
|
< 0.0001
|
|
B²
|
853.29
|
1
|
853.29
|
3071.69
|
< 0.0001
|
|
C²
|
0.0676
|
1
|
0.0676
|
0.2435
|
0.6368
|
|
Residual
|
1.94
|
7
|
0.2778
|
|
|
|
Lack of Fit
|
0.9658
|
3
|
0.3219
|
1.32
|
0.3857
|
not significant
|
Pure Error
|
0.9787
|
4
|
0.2447
|
|
|
|
Cor Total
|
1353.13
|
16
|
|
|
|
|
Adjusted R²=0.9967, R²=0.9986, Predicted R²=0.9874, Adequate Precision = 74.7984
|
• A = Adsorbent dose B = pH C = Contact time |
Figure 7 presents Pareto diagram related to the main effects of the variables and the interaction of the amount of pesticide absorbed by the Co/Ni/[email protected] composite, results are consistent with the ANOVA results.
The p-value for non-compliance in this design for diazinon removal is < 0.0001, which specifies the suitability of this model. The polynomial that expresses the relationship between the answer and the effective sentences is presented as Eq. (3). where A, B, and C represent the adsorbent dose, pH, and contact time, respectively.
R = + 82/30 + 6/08A + 2/10B-17/87C-1/54AB + 0/1025AC-2/16BC-4/90A2-14/24B2 + 0/1268C2 (3)
The random pattern of residual distribution diagram allows effective uncontrollable factors to be detected during the experiments. The random and two-sidedness distribution of errors in each experiment indicates that the errors are not systematic and the proposed model can be defined to perform defined experiments. Figure. 8a shows the residual curve of each experiment to remove diazinon pesticide by synthesized adsorbents. As can be seen, the error in performing the experiments is a random error and is not systematic. Therefore, the prepared model is able to provide the percentage of removal of the desired pesticide from the solution. The Predicted R² of 0.9874 is in reasonable agreement with the Adjusted R² of 0.9967; i.e. the difference is less than 0.2. Also Fig. 8b indicated that the value of adequate precision is 74.789...
3.3. Response Surface
As illustrated in Figure. 9(a-c), the contour and 3D response surface plots were obtained to investigate the interaction between all factors (adsorbent dose, pH and contact time).
Figure 10 presents the values of the desirability and optimal factors affecting the response under different conditions. On the basis of above experiments, removal efficiency of 84% was obtained under following conditions: adsorbent dosage 25 mg, pH value 6.9 and contact time 12 min, which is very close to the actual value.
On the basis of the above results it can be concluded that one of the main factors influenced on the adsorption efficiency is the pH of the solution. As can be observed the pH of 7 of the efficiency adsorption reaches the maximum amount. The zpc value of Co/Ni/[email protected] is about 10 that means at the pH > 10 its surface is positively charged, on the other hand, the pKa of the diazinon is 2.6 [36] indicating that the pH value higher 2.6 of the solution it has negative form. Consequently, it can be visualized that electrostatic attraction between the Co/Ni/[email protected] and diazinon is one of the main effect for the adsorption process.
In addition, the π–π stacking interaction between the imidazole ring of ZIF-8 and the aromatic ring of organophosphate pesticide could be possible another reasons for the uptake and removal of diazinon by the nanocomposite. Moreover, the possible interaction between zinc metal in the ZIF-8 structure as Lewis acid and sulfur site of diazinon as Lewis base can influence on the adsorption process.
Results indicated that the removal efficiency increases with increasing dose of absorbent, since, the available sites are enhanced by increasing dosage of adsorbent. Contact time also is another factor that can influence on the response, obviously by increasing contact time the removal percent is increased however, after a given time it has no impact. In this experiment, the optimized values of adsorbent dosage, pH and contact time were obtained 24.85 mgr, 6.95 and 11.88 min.