3.1. Photocatalyst characterization
XRD analysis was used to determine the shape of the crystals in the photocatalyst and its support. Figure 2 presents XRD diagrams of modified Iranian clinoptilolite and iron oxide supported on modified Iranian clinoptilolite. As shown in this figure, there are a large number of peaks in the diagram of the modified Iranian clinoptilolite that indicate the presence of silicon oxide crystals (JCPDS card no. 01-076-0939 and JCPDS card no. 00-044-0696). The peaks for JCPDS card no 01-076-0939 in 2q are equal to 22.1, 28.5, 36.4, and 65.3 and for JCPDS card no 00-044-0696 in 2q are equal to 10.3, 11.6, 17.6, 22.8, and 23.2. In addition, calcium silicate crystals (JCPDS card no 01-087-1260) in 2q are equal to 30.5, 32.3 and 57.6 and calcium magnesium silicate crystals (JCPDS card no 01-083-1819) in 2q are equal to 13.6, 26.5, 30.5, 49.2, 62.5, and 74.9. With loading ferrous sulfate heptahydrate on the modified Iranian clinoptilolite, the crystals of iron sulfate (JCPDS card no 01-073-0148) and sodium aluminum silicate (JCPDS card no 01-076-0898) at 2q are simultaneously equal to 14.9, 20.3 , 21.8, 23.8, 24.7, 26.7, 29.8, 31.3, 32.6, 36.0, 42.7, 47.0, 50.8, 53.5, 59.2, 61.8, 64.8 and 77.4. XRD diagrams show a similar trend in the peaks in the MIC and the FeO/MIC photocatalyst, and new peaks are formed due to the presence of Fe and Fe2+ in the FeO/MIC photocatalyst.
SEM analysis was used to evaluate the morphological and surface characteristics of MIC and FeO/MIC (Figure 3). As presented in Figure 3-a, MIC crystals have layered or a layer-like structure. The SEM image of the FeO/MIC sample (Figure 3-b) also presents the same MIC structure. As a result, MIC crystals are not affected by Fe2+ and FeO loading.
The EDAX spectrum of the FeO/MIC sample is also presented in Figure 4. Based on the results of EDAX analysis, the percentage of elements were C: 3.45 wt. %, O: 52.80 wt. %, Na: 0.18 wt. %, Mg: 0.29 wt. %, Al: 2.38wt. %, Si: 23.56 wt. %, S: 6.64 wt. %, K: 0.59 wt. % and Fe: 10.11 wt. %.
The analysis of FT-IR spectra is a very important technique for determining the characteristics of functional groups as well as the changes of these groups in photocatalysts. FT-IR spectrum of the modified Iranian clinoptilolite is presented in Figure 5-a and the spectrum of iron oxide supported on modified Iranian clinoptilolite is presented in Figure 5-b.
Figure 5 presents the main groups at peaks of 471, 611, 798, 952, 1090, 1632, and 3438 cm-1 in the MIC. The group at 472 cm-1 is related to the stretching vibrations of the Al-O bond, while the peaks at 798 and 1090 cm-1 are related to the Si-O-Si bond (Chávez et al. 2010). Vibration bonds located at 1632 and 3438 cm-1 indicate the tensile frequency of the hydroxyl functional group (OH) (Han et al. 2010). Figure 5 also presents the FT-IR spectrum of the FeO/MIC sample used to investigate changes due to the presence of FeO in the modified Iranian clinoptilolite. Breck showed that with inserting metal cations into a zeolite structure, some minor changes may occur in the position of the peaks located to the right of the tensile bond of TOT or OTO (T = Al or Si) (Breck 1984). Minor variations in peaks can be observed at 475, 623, 794, 1112, 1637, and 3455 cm-1 in the FeO/MIC spectrum. There is also a peak at 661 cm-1, indicating the presence of FeO in the MIC structure. Similar results have been observed in previous studies that used a combination of some metal cations and sulfides or oxides related to different zeolites (Nezamzadeh-Ejhieh and Khorsandi 2010; Nezamzadeh-Ejhieh and Hushmandrad 2010; Nezamzadeh-Ejhieh and Salimi 2010).
3.2. Effect of operational variables on ibuprofen in the wastewater
In this study, the effect of different variables including the initial ibuprofen concentration, photocatalyst concentration (FeO/MIC), and process time on the removal of ibuprofen in the wastewater was investigated. The initial concentration of ibuprofen (Cd,i) was studied in the range of 5-25 mg/L, the concentration of photocatalyst (CFeO/MIC) in the rang of 100-300 g/L, and the process time (t) in the range of 10-240 min.
Figure 6 shows the percentage curves of ibuprofen removal over time at different photocatalyst concentrations (FeO/MIC). As shown in the diagrams in this figure, the amount of ibuprofen in the wastewater decreased over time. With the passage of time up to 120 min, the slope of ibuprofen removal decreased, but the overall percentage of ibuprofen removal increased. At 150 min, the percentage of ibuprofen removal was almost constant over time, but until the end of 240 min, the dye removal rate was slightly increased.
As can be observed in the diagrams in Figure 6, the concentration of the photocatalyst plays an important role in the rate of ibuprofen removal. With increasing the concentration of the photocatalyst from 100 mg/L to 300 mg/L, there was an increasing trend in the removal of ibuprofen in the wastewater containing this drug. In addition, comparing Figures 6-a, 6-b, and 6-c, it is clear that with increasing the initial concentration from 5 mg/L to 15 mg/L (at all the photocatalyst concentrations), there was a relatively proper increasing trend in the removal of ibuprofen from the wastewater containing this drug. With increasing the initial concentration of ibuprofen from 15 mg/L to 25 mg/L, the increasing trend of ibuprofen removal continued, but the rate of increase in ibuprofen removal was lower than the state when the initial concentration changed from 5 mg/L to 15 mg/L.
According to the results presented in Figures 6-a, 6-b and 6-c, the initial concentration of 25 mg/L, photocatalyst concentration of 300 mg/L, and the process time of 210 min can be introduced as the optimal operating conditions. Under the mentioned condition, the removal rate of ibuprofen is 99.80%.
3.3. Kinetic modeling of photocatalytic process
Normally, the kinetics of photocatalytic reactions are consistent with the Langmuir-Hinshelwood model (Faramarzpour et al. 2009; Zhang et al. 2012):
where is the pollutant removal rate in mg/(L.min) after the time t in min, C is the pollutant concentration in mg/L, k is the reaction rate constant in mg/(L.min), and is the constant of Langmuir-Hinshelwood absorption equilibrium in L/mg. When the initial concentration (C0) is about mM ( C0 is very small), the equation can be simplified to a quasi-first-order equation as follows:
where k'=kK in min-1 is a quasi-first-order reaction constant, and its value is obtained through plotting the diagram of in terms of time.
Figure 7 indicates that has a linear relationship with t, which is proved by a high correlation coefficient (R2>0.95). Table 2 presents the constant values of quasi-first-order reaction (k') for different conditions (different values of the initial ibuprofen concentration and the photocatalyst concentration) in the ibuprofen removal photocatalytic reaction. As shown in this table, with increasing the initial concentration of ibuprofen as well as increasing the photocatalyst used, the value of the constant of reaction increases. A similar trend was observed in a study by Mousavi Mortazavi et al. that investigated the elimination of furfural (Mousavi-Mortazavi and Nezamzadeh-Ejhieh 2016).
3.4. Recovery and reusability of FeO/MIC
The process of photocatalyst recovery was performed according to the mentioned method under optimal conditions (the initial ibuprofen concentration of 25 mg/L and the adsorbent concentration of 300 mg/l and the process time of 210 min). In order to recover the magnetic photocatalyst (FeO/MIC), a strong magnet was used first to separate FeO/MIC from the refined sample. The isolated FeO/MIC was then washed several times with deionized water and after drying, it was used again as a photocatalyst in the next step. Figure 8 presents a summary of the results of tests on photocatalyst recovery. As presented in this figure (Figure 8), after re-use of the photocatalyst for five times, the removal rate of ibuprofen was still above 91%, and finally, in the sixth and seventh re-use, the ibuprofen removal rate was decreased to about 83% and 70%, respectively. These results indicate the high potential of the photocatalyst used in this study during the recovery process.