3.1 The effect of ultrasonic frequency and intensity on the production of H 2 O 2
To figure out whether the amount of sono-generated H2O2 is sufficient to create the extra in situ Fenton reaction in sonication, herein, quantification of H2O2 was conducted at different ultrasonic frequencies and intensities. The effect of ultrasonic frequency and intensity on the production of H2O2 was shown in Fig. 1 (a) and (b). Concentration vs. time profiles were therefore analyzed with a simple pseudo-zero-order equation as follows:
$$Ln{\left[{H}_{2}{O}_{2}\right]}_{t}/{\left[{H}_{2}{O}_{2}\right]}_{0}=-{\text{k}}_{\text{a}\text{p}\text{p},1}\text{t}$$
3
Where kapp,1 is the apparent zero-order rate constant of H2O2 formation, t is the irradiation time, and [H2O2]0 and [H2O2]t are the concentrations at time = 0 and time = t, respectively.
As illustrated in Fig. 1 (a) and (b), the formation of H2O2 followed pseudo-zero-order kinetics at each ultrasonic frequency and intensity (R2 = 0.928). Meanwhile, the H2O2 yields at different frequencies followed the order: 619 > 406 > 226 > 800 > 100 kHz. Of note, the •OH yields were reported in the order of 354 > 620 > 803 > 206 > 1062 kHz, which is different from that of H2O2 herein (Yang, Sostaric, Rathman, & Weavers, 2008). That the H2O2 production at 226 kHz is higher than that at 800 kHz can be attributed to more •OH self-recombination events happening at the water-bubble interface per unit time due to the longer ultrasonic cycle time despite lower •OH production. However,the self-recombination rate constant of •OH under ambient condition was measured to be 5.5×109 mol L − 1 s− 1, which indicated that the self-recombination reaction rate was controlled by the mass transfer to some extent. From this viewpoint, an appropriate higher frequency is beneficial to the H2O2 production due to greater effects of diffusion from the internal bubble to the water-bubble interface (Campbell, Vecitis, Mader, & Hoffmann, 2009).
The effect of ultrasonic intensity on the production of H2O2 with time is shown in Fig. 1(b). kapp,1 proportionally increases from 0.136 min− 1 to 2.762 min− 1 with an increase in lower ultrasonic intensity from 0.20 to 0.69 W/cm2 (kapp,1= 6.575[ultrasonic intensity] -1.34, R2 = 0.973). These results indicated that H2O2 production is strongly dependent on the ultrasonic frequency and intensity under air atmosphere condition. Moreover, compared with ultrasonic frequency, the intensity seems to have a more significant impact on the production of H2O2. The increase in ultrasonic intensity would increase the mixing intensity, in addition to yielding higher numbers of cavitation bubbles(Price & Lenz, 1993) and hence higher yields of •OH. The optimal ultrasonic frequency for dimethoate degradation was found to be 610 kHz, which is almost close to that for H2O2 production according to our previous study(Yao et al., 2011). It is proved that the ultrasonic frequency of 610 kHz is beneficial to the degradation of dimethoate and the formation of H2O2 in bulk solution.
Furthermore, the degradation efficiency of sono-generated H2O2 on dimethoate was also verified. Considered totally 160 µM H2O2 was generated at the ultrasonic frequency of 619kHz and power density of 0.69 W/cm2 after 60 minutes, the degradation of dimethoate by 160 µM H2O2 was conducted. As seen in Fig. 2, only 6.0% was degraded within 120 minutes. Compared with single sonication(Yao et al., 2011), sono-generated H2O2 had almost little effect on the degradation of dimethoate, which can be explained by the •OH produced by sonication plays a major role in the degradation of dimethoate.Therefore, it could be concluded that the sono-generated H2O2 does not play a significant role in the overall sono-degradation of dimethoate. It’s more desired to use the sono-generated H2O2 from an economic point of view.
3.2 The effect of ultrasonic intensity on the decrease of pH value
Since conventional Fenton reactions are always sensitive to solution pH, different initial pH values and ultrasonic intensities were used to verify whether sonication can produce favorable acidic environmental conditions for the Fenton reaction. The results were shown in Fig. 3. It is observed that the pH value vs. time curves under the different ultrasonic intensities kept a similar decline profile in which the pH value dropped sharply during the initial 5 min and then slightly to 3.5 to 4.1 in the next 25 minutes at the fixed frequency of 600 kHz and same initial pH value of 7.0. Even at a relatively high initial pH value of 9.0, the pH value can still decrease to 4.1 after 30 minutes of sonication with the ultrasonic intensity of 0.69 W/cm2.
It was proved that the sonication can create an acid condition which quite approaches the optimal Fenton reaction pH value of 2.8 ~ 3.0 in initial neutral aqueous solutions (Eqs. (4)–(8)) (Brillas, Sirés, & Oturan, 2009).
3.3 The effect of ferrous ions dosage on the consumption of H 2 O 2
The variation of H2O2 concentrations were monitored to examine the utilization of Fe2+ in this sono-Fenton system. It was found that the addition of Fe2+ into the bulk solutions would consume the H2O2 formed by sonication as shown in Fugure 4. Considered the Fenton reaction as follows (Eqs. (9) to (15)), it was obvious that at the beginning of the sonication, the amounts of H2O2 was insufficient for the reduction of Fe3+ (Eqs. 12). However, the significant enhancement of dimethoate degradation gives evidence that Fenton reaction between Fe2+ and H2O2 then yields •OH can keep going,which indicated that the free radicals , and •H generated in the ultrasonic irradiation, as shown in reaction (4), (7) and (8), should play a significant role in the reduction of Fe3+.
k9 = 63 M− 1s− 1 (9)
k10 = 3 × 107 M-1s-1 (10)
k11 = 4.3 × 108 M-1s-1 (11)
k12 ≤ 3 × 10-3M-1s-1 (12)
k13 = 1 × 103 M-1s-1 (13)
k14 = 1.5×108 M-1s-1 (14)
•H + Fe3+→Fe2+ + H+ (15)
The effect of Fe2+ concentration on the amount of H2O2 in solution at pH = 3.0 and pH = 9.0 as a function of irradiation time was illustrated in Fig. 4, respectively. When the initial pH value was 3.0, the effect of different iron ion dosages on the net yields of H2O2 were obviously different. The concentration of H2O2 in solution decreases substantially with increased Fe2+ concentration, as expected. The concentration of H2O2 is significantly reduced in the presence of relatively low concentrations of Fe2+ indicating the conversion of H2O2 to •OH was enhanced with the addition of Fe2+. When the initial pH value is 9.0, the different iron ion dosage does not affect the net yield of H2O2, which may be due to the inhibition of Fenton reaction by the initial alkaline conditions. Furthermore, our previous research has proved that the maximum degradation rate was obtained with the Fe2+ concentrations at 100 µM in initial neutral condition. A slight decrease in the degradation rates was observed at Fe2+ concentrations over 100 µM due to the direct scavenging effect of the •OH by Fe2+. Although it has been proved in the previous discussion that with the progress of the reaction, the pH environment of the optimal reaction conditions will eventually be achieved, the different forms of Fe2+/Fe3+ at different initial pH values determine the change of the optimal addition dosage of Fe2+ for both dimethoate degradation and omethoate controlling. Hence,the dosage of Fe2+ less than or equal to 300µM was selected to explore the optimal addition dosage of Fe2+ under different initial pH values.
3.4 The interaction effects of ultrasonic intensities, initial pH values, and Fe 2+ dosages on the Fe2+ enhanced sono-degradation of dimethoate
The response surface methodology with the BBD model was utilized to simulate the interaction effects of ultrasonic intensities, initial pH values and Fe2+ dosages on the sono-degradation of dimethoate. BBD and its experiment results are shown in Table 2. (Degradation curves of dimethoate under 15 working conditions were shown in figure S1.)
Table 2. BBD and its experiment results
Run order
|
Code
|
kapp/min-1
|
X1
|
X2
|
X3
|
Fitted by Experimental results
|
R2
|
1
|
0
|
0
|
0
|
0.1786
|
0.991
|
2
|
1
|
-1
|
0
|
0.1291
|
0.992
|
3
|
0
|
1
|
-1
|
0.2713
|
0.988
|
4
|
0
|
1
|
1
|
0.0937
|
0.982
|
5
|
1
|
1
|
0
|
0.2336
|
0.990
|
6
|
-1
|
0
|
-1
|
0.1905
|
0.990
|
7
|
-1
|
0
|
1
|
0.0756
|
0.995
|
8
|
0
|
-1
|
-1
|
0.0976
|
0.987
|
9
|
-1
|
-1
|
0
|
0.0407
|
0.994
|
10
|
-1
|
1
|
0
|
0.0276
|
0.991
|
11
|
0
|
0
|
0
|
0.1766
|
0.995
|
12
|
0
|
0
|
0
|
0.1676
|
0.989
|
13
|
1
|
0
|
1
|
0.2467
|
0.991
|
14
|
0
|
-1
|
1
|
0.1108
|
0.994
|
15
|
1
|
0
|
-1
|
0.3904
|
0.990
|
Experimental data were analyzed using Design-Expert program 8.0 trial version including ANOVA to obtain the interaction between the variables and the response. Two-dimensional contour plots and three-dimensional curves of the response surfaces were developed using the Origen 8.0. Optimization of the Y was carried out by the Design-Expert program 8.0 trial version.
An empirical relationship between the response and independent variables (in the term of coded) was attained and expressed by the following second-order polynomial equation:
Y1 = 0.17940 + 0.08317 X1 + 0.03100 X2 − 0.05289 X3 − 0.0753 X22 + 0.0428 X32 + 0.0294 X1X2 − 0.0477 X2X3 (16)
Of note,the quadratic terms (X12 )and the interaction terms X1X3 was found to be not significant by the analysis of variance (ANOVA) and thus be ignored in Eq. (16). Positive and negative signs in front of the terms in Eq. (16) indicated the synergistic effect and antagonistic effect, respectively.
The model F-value of 44.18 implied that the model was significant for the pseudo-first-order degradation constant of dimethoate and there was only a 0.01% chance that a model F-value of this large could occur due to noise. Besides, the model adequate precision of 22.844 (greater than 4 as usually desired) that was greater indicated that an adequate signal for the model was used to navigate the design space. The "Lack of Fit F-value" of 16.42 implies there is a 5.85% chance that a "Lack of Fit F-value" this large could occur due to noise.
The values of "Prob > F" less than 0.0500 indicate model terms are significant, while the values greater than 0.1000 indicate the model terms are not significant. In this case, the terms of individual variables terms (X1, X2, and X3), quadratic terms (X22 and X32), and interaction terms (X1X2 and X2X3) were confirmed to be significant.
Table 3 summarizes the analysis of variance (ANOVA) results. The Model adjust-R2 value of 0.956 was relatively high, which indicated that the model obtained was able to give a convincingly good estimate of response in the studied range.
Table 3. The Analysis of variance (ANOVA) for the quadratic model
Source
|
Freedom degrees
|
Adjusted Squares sum
|
Adjusted
|
F-Value
|
P-Value
|
|
Model
|
7
|
0.127624
|
0.018232
|
44.18
|
0.000
|
significant
|
Linear
|
3
|
0.085402
|
0.028467
|
68.99
|
0.000
|
|
Ultrasonic intensity
|
1
|
0.055341
|
0.055341
|
134.12
|
0.000
|
|
Fe2+ dosage
|
1
|
0.007686
|
0.007686
|
18.63
|
0.003
|
|
Initial pH value
|
1
|
0.022376
|
0.022376
|
54.23
|
0.000
|
|
Square
|
2
|
0.029661
|
0.014831
|
35.94
|
0.000
|
|
Fe2+ dosage* Fe2+ dosage
|
1
|
0.020948
|
0.020948
|
50.77
|
0.000
|
|
Initial pH value*Initial pH value
|
1
|
0.006851
|
0.006851
|
16.60
|
0.005
|
|
2-Way Interaction
|
2
|
0.012561
|
0.006281
|
15.22
|
0.003
|
|
Ultrasonic intensity* Fe2+ dosage
|
1
|
0.003453
|
0.003453
|
8.37
|
0.023
|
|
Fe2+ dosage*Initial pH value
|
1
|
0.009108
|
0.009108
|
22.07
|
0.002
|
|
Error
|
7
|
0.002888
|
0.000413
|
|
|
|
Lack-of-Fit
|
5
|
0.002820
|
0.000564
|
16.43
|
0.058
|
Not significant
|
Pure Error
|
2
|
0.000069
|
0.000034
|
|
|
|
Total
|
14
|
0.130513
|
|
|
|
|
R-sq R-sq(adj) R-sq(pred)
97.79% 95.57% 84.25%
Figure 5 shows the actual experimentalvalues versus predicted output values for the pseudo-first-orderdegradation constant of dimethoate. From Fig. 5, it is noted that a high correlation between the experimental and predicted values, showing that the model was well fitted and had a good prediction ability for the pseudo-first-order degradation constant of dimethoate.
From ANOVA results, significant interaction terms were found to exist between the main factors (X1X2, X2X3). Figure 6 shows contour response surface plot of pseudo-first-order degradation constant Y versus Fe2+ dosage X1 and ultrasonic intensity X2 at initial pH value X3 of -1, 0 and 1 to investigate the interactive effects. Figure 7 shows contour response surface plot of pseudo-first-order degradation constant Y versus Fe2+ dosage X1 and initial pH value X3 at ultrasonic intensity X2 of -1, 0 and 1 to investigate the interactive effects. It was observed that an optimal Fe2+ dosage can be found at each ultrasonic intensity and the optimal Fe2+ dosage increased with increasing of the ultrasonic intensity, while deceased as the initial pH value increases.
In conclusion, the initial pH = 3, Fe2+ dosage = 200 µM, ultrasound intensity = 0.69 W/cm2 were the best reaction condition of the simple sono-Fenton system. The results were shown in Fig. 8 that the optimum degradation efficiency in this sono-Fenton system was 95.3% within 10 minutes, which is 38.1% higher than that of single sonication. It was calculated that 80 µM H2O2 will be produced in bulk solution under 20 min sonication, which was used to figure out the optimal Fe2+/H2O2 ratio in the simple sono-Fenton system. Of note, the calculated H2O2 concentration was just an approximate steady-state concentration that resulted from a dynamic circulation. The ratio of the optimal Fe2+/H2O2 in Fenton system was calculated by the following equation:
$$\text{T}\text{h}\text{e} \text{o}\text{p}\text{t}\text{i}\text{m}\text{a}\text{l} {Fe}^{2+}/{H}_{2}{O}_{2} \text{r}\text{a}\text{t}\text{i}\text{o}={C}_{{Fe}^{2+}}/{C}_{H2O2}$$
17
Where CFe2+ is the concentration of Fe2+ in the system at the optimal efficiency, CH2O2 is the concentration of H2O2 in the system at the optimal efficiency, respectively.
The optimal ratio in this simple sono-Fenton system was 25 times higher than that of the conventional Fenton reaction( the optimal Fe2+/H2O2 ratio = 0.1)(Chen et al., 2011),which indicated that the Fe3+/Fe2+ cycle might significantly be accelerated and the concentration of Fe2+ was always sufficient in this simple sono-Fenton system. The accumulation of Fe3+ is the rate-limiting step of conventional Fenton reaction(Chen et al., 2011), but it is not suitable for this simple sono-Fenton system. It seems that the sono-generated H2O2 contributes to the continuous formation of the oxidizing species(•H), which led to accelerating the reduction of Fe3+. Meanwhile, the mass transfer generated by sonication might also accelerated the Fe3+/Fe2+ cycle. In this simple sono-Fenton system, the addition of trace Fe2+ could react with the sono-generated H2O2 to generate •OH and kept the oxidation reactivity in the bulk solution.
Based on the above findings, with the progress in this simple sono-Fenton system, without adjusting the initial pH value, the reaction can also be carried out at the appropriate pH value. In practical application, the ratio of acid consumption at different initial pH values was calculated by the following equation:
$$\text{A}\text{c}\text{i}\text{d} \text{c}\text{o}\text{n}\text{s}\text{u}\text{m}\text{p}\text{t}\text{i}\text{o}\text{n} \text{r}\text{a}\text{t}\text{i}\text{o}=({10}^{-3}-{10}^{-a})/({10}^{-3}-{10}^{-b})$$
18
Where a is the initial pH of the real water body, b is the pH value of the simple sono-Fenton system after 5 minutes of sonication, respectively.
When pH = 3.0 was adjusted from the real water body to be treated which the initial pH = 7.0, the H+ required by conventional Fenton reaction is 1.1 times of this simple sono-Fenton reaction. This proved that the pH adjustment cost of this simple sono-Fenton system is greatly reduced by the acidic condition formed after 5 minutes of sonication.
In the conventional Fenton system, when the initial pH value is neutral or alkaline, Fe2+ tends to form amorphous precipitates, which makes the reaction activity in the system decrease and is not conducive to the formation of •OH(Wang & Liu, 2014). However, an increase in acidity caused by sonication provides an ideal condition for this simple sono-Fenton system, it not only saved the cost of initial pH adjustment but also made the reaction always in high activity. It was proved that the synergistic effect of ultrasound and Fenton reaction can be induced by utilizing sono-generated H2O2 with trace Fe2 + addition.
3.5 Control of dimethoate intermediates by simple sono-Fenton system
In previous studies, the primary degradation pathway of dimethoate under sonication was discussed(Yao et al., 2011). In this study, to verify the control of this simple sono-Fenton system on the intermediate products of dimethoate, and explore whether the degradation of omethoate is effective after adding trace Fe2+, sonication of dimethoate was conducted under the optimal conditions of this simple sono-Fenton system and single sonication. Nine intermediates for dimethoate have been identified using GC-MS detection as shown in Table 4. (The mass spectra of dimethoate (DIM) and its degradation Intermediates (No.1–No.9) formed during the sono-degradation in the presence of Fe2+ (200µM) was shown in figure S2). As shown in Fig. 9 (a), in this simple sono-Fenton system, dimethoate degradation intermediates mainly were dimethyl phosphite after 45 minutes. And the concentrations of other products were all below the detection line. Compared with the degradation intermediates of sono-degradation alone, Fig. 9 (b), the amount of N-(methyl) mercaptoacetylamide、O, O, S-trimethyl phosphorothioate, 2-(methyldisulfanyl) acetamide, and Omethoate of later decreased obviously in this simple sono-Fenton system. This proves that the system can improve the degradation efficiency of hydrophobic intermediates (The hydrophobic and hydrophilic coefficients of intermediates were shown in Table. S1), and had good control on Omethoate with high toxicity.
In previous studies, it has been known that single sonication on dimethoate mainly depends on •OH formed by cavitation bubble collapse(Yao et al., 2011). The attack of •OH on P = S resulted in the formation of omethoate and cyclic hexatomic sulfur, but they decreased significantly under this simple sono-Fenton system. Therefore, it can be inferred from the above degradation pathways that P-S may be an important attack point for •OH after the addition of Fe2+. In order to better explore the control effect of this simple sono-Fenton system on toxic by-products, based on the reason that it is difficult to determine TOC at a low concentration of dimethoate, the change of concentration of ionic intermediates (HCOO-, PO43-, SO42-) were used for verification. It can be seen from Fig. 10 that the formation rate of the three ions in the solution was doubled after adding a trace Fe2+, which indicated that this simple sono-Fenton system had a great control effect on the dephosphorization and desulfurization of substances. As discussed above, the sonication was more capable to degrade compounds with higher hydrophobicity so that the degradation intermediates which always showed higher hydrophilicity compared to the parent compound would easily accumulate. However, this simple sono-Fenton system has the advantage to improve the degradation of the accumulated intermediates since the Fenton reaction occurred in the bulk aqueous phase rather than in the interface of cavitation bubbles. The use of sono- generated H2O2 solved the secondary increase of omethoate toxicity caused by the accumulation of ineffective H2O2(Inoue, Okada, Sakurai, & Sakakibara, 2006). In addition, the micro-mixing effect by turbulent flow in the reactor during sonication would result in the acceleration of Fenten reaction due to the increase in the mass transfer rate induced from hydrodynamic cavitation. Based on the above findings that this simple sono-Fenton system was more effective to improve the degradation in bulk solution.
Table 4. GC-MS-EI retention times (Rt) and spectral characteristics of dimethoate identified intermediates
No.
|
Intermediates
|
Retention times(min)
|
Similarity (%)
|
Reference ions
|
Characteristic ions(m/z)
|
1
|
Dimethyl phosphite
|
4.333
|
97
|
109
|
109,95,80,77,65,51,47,45
|
2
|
O, O, O-trimethyl thiophosphate
|
7.417
|
90
|
93
|
156,126,93,79,63,47
|
3
|
N-(methyl) mercaptoacetamide
|
9.275
|
80
|
58
|
105,78,72,58,48
|
4
|
O, O, S-trimethyl phosphorothioate
|
9.783
|
93
|
156
|
140,110.109,95,79,65,47
|
5
|
2-(methyldisulfanyl) acetamide
|
10.108
|
80
|
58
|
119,73,62,58,47
|
6
|
O, O, S-trimethyl phosphorodithioate
|
11.125
|
94
|
125
|
172,157,141,125,109,93,79,77,63,47
|
7
|
O, S, S-trimethyl phosphorodithioate
|
13.217
|
86
|
172
|
172,157,141,125,94,79,62,47, 45
|
8
|
Cyclic hexatomic sulfur
|
16.858
|
82
|
128
|
206,174,160,142,128,110,96,78,64,45
|
9
|
Omethoate
|
18.142
|
100
|
110
|
156,141,126,110,95,79,58,47,44
|
3.6The simple sono-Fenton system in the real water sample
In order to verify the practicability of this simple sono-Fenton system, explore whether the degradation efficiency of it is affected in the real water samples. The removal effect of dimethoate by sonication after adding ferrous ion into pure water, lake water, and tap water were compared (The water quality of real water samples were shown in Table. 1S). It is observed from Fig. 11 that dimethoate concentration vs. time curves under the different water samples kept a similar decline profile, and dimethoate was removed to a lower concentration level within 10 minutes. The experimental results showed that the removal efficiency of dimethoate by this simple sono-Fenton system was not affected by the water environment, and had a wide range of applicability.