Optimization of multiple parameters for treatment of coking wastewater using Fenton oxidation

2.1 Materials

The CWW was collected from Steel Authority of India Limited (SAIL) plant situated in Bhilai, Chhattisgarh, India. Analysis of CWW before and after treatment is presented in Table 1·H₂O₂ (30% W/W), ferrous sulphate heptahydrate (FeSO₄·7H₂O), sulphuric acid, sodium hydroxide and activated carbon used were made of Merck Ltd. All the chemicals used were analytical and laboratory grade and the solutions were prepared using deionised water.

2.2 Analytical procedure

COD was determined by the standard dichromate closed reflux method. Phenol was determined by colorimetric method using 4 amino anti pyrine and potassium ferri cyanide (Rice et al., 2012). Analysis of cyanide was done using Merck standard kit. Hardness, chlorine was determined by standard method (Rice et al., 2012). Color reduction was estimated by relative color content of treated and untreated effluent. For this the relative adsorption of wastewater was measured at 475 nm using spectrophotometer (Chaudhari et al., 2005).

2.3 Experimental setup and procedure

Fenton process was performed in a 0.5 dm³ beaker placed on a magnetic stirrer platform. 250 ml of CWW was taken in it and the operating pH was maintained by the addition of dilute sulphuric acid, after that ferrous sulphate was added to the beaker and then desired amount of H₂O₂ was injected. The treated samples were taken out at certain time intervals and analysed for pollutants contain. Whenever desired, the experimental runs were repeated to check the accuracy of the results. More than 3% deviation in results were further repeated to check its reliability. The starting time of the reaction temperature was considered as zero time. The percentage reduction of COD, phenol and cyanide were calculated by using equation;

Adsorption treatment of Fenton treated CWW (COD_i $\cong$ 590.0 mg/L, phenol_i = 39.49 mg/L and cyanide_i = 5.2 mg/L) was carried out in a water bath shaker in batch mode. 50 ml of Fenton treated CWW (at optimum condition) was taken in a 100 ml conical flask. The required amount of adsorbent was added and kept in shaker for proper contact of adsorbent and pollutants contained in CWW. At different time intervals about 5 ml samples were taken out and adsorbents were allowed to settle down. The supernatant was then analysed for COD and phenol content. Effect of pH, adsorbent dosage and adsorption time were taken as experimental parameters for the COD and phenol removal. The cyanide and other parameters were also determined at the end of the experiment.

2.4 Experimental design and optimization

RSM was used to design the experiments of total 20 runs to develop the quadratic model, and to evaluate the individual and interaction effects of the parameters on the COD and phenol removal efficiency. RSM has been established by Box and Wilson and it is well designed for fitting a quadratic surface, which generally works well for the process optimization. The experimental data were analysed using Minitab and the regression models were developed using three independent variables including t_R = 0–2 h, Cw = 0.1–3 g/L, H₂O₂ = 0.07 M-0.36 M. For statistical analysis, the three independent variables were coded as X1 (t_R), X2 (Cw), X3 (H₂O₂). The result obtained from the RSM through central composite design (CCD) model was written in the form of following equation:

3.1 Effect of pH

pH of solution play an important role in Fenton treatment of wastewater, therefore, in present studies its effects were tested on COD, phenol and cyanide reduction. The effect of pH_i on removal efficiency of COD and phenols are shown in Fig. 1a, b, which indicates the removal efficiency was highest at pH = 3 with the maximum reduction of COD = 78.57% and phenol = 86.04% in one hour. Other parameters were kept constant to H₂O₂ = 0.3 M, FeSO₄ = 3 g/L, T = 30 °C. At pH = 3, 4, 5, 6, 7 and 9, respectively, 78.57, 68.32, 65.98, 48.76, 48.12 and 47.42% COD reduction; and 86.04, 79.08,72.65, 69.54, 66.61 and 64.71% phenol reductions were achieved in 1 h. After 1 h, COD and phenol reductions became highly slow. As the pH_i was decreased from the effluent pH = 9, the percentage reduction efficiency of COD, phenol and cyanide was increased continuously. At pH_i = 10 negligible degradation of organic matters were observed. As result reflects that with the increase in pH_i from 3 to 9, COD reduction was decreased to 47.42% from 78.57 and phenol reduction was decreased to 64.71% from 86.04 in 1 h. The reason is, when pH_i is increases; Fe²⁺ starts to precipitate under alkaline conditions, and its availability for reaction decreases. Apart from this, hydrogen peroxide is also not stable in alkaline condition and it breaks down into oxygen and water and loses its oxidative capacity (Sanja et al., 2008). At extreme low pH (<3), hydrogen peroxide are more stable and slowly decomposes to produce hydroxyl radicals. Due to these, Fe²⁺ and H₂O₂ have difficult to form an effective redox reaction under high and very low pH. Moreover, low initial optimum pH (3–4) ensures greater yield of hydroxyl radicals because of no inhibiting effect of iron complexes in this pH range (Deng and Englehardt, 2006; Libing et al., 2012). Fenton process for the treatment of CWW was found to more efficient at pH = 3. An acidic media is better for the dissolving of Fe²⁺, which gives the large amount of hydroxyl radicals by the introduction of H₂O₂ in CWW for the degradation of organic pollutants.

Close

Fig. 1 Effect of pH on (a) reduction of COD (b) reduction of phenol during Fenton oxidation process·H₂O₂ = 0.3 M, FeSO₄ = 3 g/L (0.043 M), COD_i = 2810.0 mg/L, Phenol_i = 283.0 mg/L, T = 30 °C.

Export to PPT

3.2 Effect of oxidant concentration

The oxidant H₂O₂ plays an important role in Fenton process because it generates the hydroxyl radicals. Treatments of CWW was carried out using different amount of H₂O₂ keeping the parameters T = 30 °C, t_R = 2 h and Cw = 3 g/L(0.043 M) as constant. As the H₂O₂ concentration was increased the degradation or percentage reduction of COD and phenol was also increased. The H₂O₂ concentration was taken in the range of 0.07 M-0.36 M. The degradation rate of organic matters was rapid in initial 10 min, after that it slows down upto 60 min and becomes almost constant with further increase in time. The percentage reduction of COD achieved was 71.42, 73.21, 76.00, 78.57 and 83.66%, respectively while phenol was 62.32, 71.53, 78.98, 86.04 and 87.46%, respectively, with H₂O₂ doses of 0.07 M, 0.14 M, 0.22 M, 0.30 M, 0.36 M. This molar concentration is equal to 2.5, 5.0, 7.5, 10.5 and 12.5 g/L H₂O_2, respectively. The results are shown in Fig. 2a, b. As the H₂O₂ dose increases, formation of hydroxyl radicals also increases and decreases slowly with the rise in the pH of the solution (Libing et al., 2012). Further addition of H₂O₂ to desired amount leads to scavenging effect on hydroxyl radicals, thereby reducing usage of hydroxyl radicals (Wang et al., 2014). Therefore, it is very important to select an optimum concentration of H₂O₂. In our experiment, desirable dosage of H₂O₂ found to 0.3 M, however there was no significant enhancement in pollutant reduction when H₂O₂ amount was increased from 0.3 M- to 0.36 M.Fig. 3.

Close

Fig. 2 Effect of H₂O₂ amount on percentage reduction of (a) COD and (b) phenol during Fenton oxidation FeSO₄ = 3 g/L (0.043 M), COD_i = 2810.0 mg/L, Phenol_i = 283.0 mg/L, T = 30 °C.

Export to PPT

Close

Fig. 3 Effect of catalyst mass loading on (a) COD reduction (b) phenol reduction during Fenton oxidation·H₂O₂ = 0.3 M, COD_i = 2810.0 mg/L, Phenol_i = 283.0 mg/L, T = 30 ^oCFeSO₄: 0.1, 0.25, 0,75, 1.5, and 3 g/L = 0.00144, 0.0036, 0.0108, 0.0216 and 0.043 M.

Export to PPT

3.3 Effect of ferrous salt dosages

The desired amount of Fe²⁺ / Fe³⁺ is necessary to get optimum results, therefore the amount of ferrous sulphate dosage was varied to 0.1, 0.25, 0.75, 1.5 and 3.0 g/L i.e 0.0014, 0.0036, 0.0108, 0.0216 and 0.043 mol Fe²⁺, respectively by keeping the value of pH = 3 and H₂O₂ = 0.3 M constant . From the experiments it was observed that with the increase in ferrous sulphate dosage, the percentage reduction of organic matter was also increased up to certain Cw, and then became steady as shown in Fig. 2ab. At ferrous sulphate dose of 0.1, 0.25, 0.75, 1.5 and 3.0 g/L, respectively, the degradation of COD was 68.57%, 71.42%, 74.51%, 76.71% and 78.57% ; and phenol was 73.98%, 77.65%, 81.43%, 84.24% and 86.04% . Fe²⁺ is necessary in Fenton process because it catalyse to generate the hydroxyl radical, without Fe²⁺, the H₂O₂ becomes harder (Wang et al., 2014).

3.4 Effect of iron hydrogen peroxide ratio

Effect of ratio of iron and hydrogen peroxide concentration was also studied by varying Fe²⁺ at H₂O₂ = 10.5 g/L (0.3 M) and pH = 3. The results are presented in Fig. 4. The increase in the reduction efficiencies of COD and phenol were observed highest when ratio of Fe²⁺/ H₂O₂ was increased from 0.009 to 0.15, after that there is no significant rise in reduction efficiencies. In Fenton process, catalyst i.e. iron is necessary element to generate the hydroxyl radicals. At low Fe²⁺, H₂O₂ becomes hard to decompose and consequently to generate the hydroxyl radicals with Fe²⁺, but when Fe²⁺ becomes high it causes enhanced H₂O₂ decomposition and thus large amount of hydroxyl radicals which causes scavenging effect (Wang et al., 2014). To avoid the scavenging effect we have to chose the optimum dosage of catalyst to increase the reaction efficiencies, which decreases the wastewater treatment cost as well as time (Chan and Chu, 2003; Guedes et al., 2003).

Close

Fig. 4 Effect of Fe²⁺/ H₂O₂ ratio (by weight) on the COD and phenol reduction·H₂O₂ = 10.5 g/L or 0.3 M, pH = 3.

Export to PPT

3.5 Effect of temperature

The effect of T in the range of 30–60 °C on the performance of Fenton process was studied by keeping pH = 3, Cw = 3 g/L (0.043 M), H₂O₂ concentration = 0.3 M and t_R = 1 h as constant. The results are presented in Fig. 5a, b. From the experimental data it was observed that as the temperature was increased the degradation of organic matters like COD and phenol was also increased marginally. In chemical reaction, the reaction rate constant (k) is increased with temperature, due to this, rate of degradation/ consumption of reactants also increases with temperature. At higher temperature the activity of Fenton process retards due to the thermal decomposition of H₂O₂ (Ghime and Ghosh, 2017). Also due to combined effect, the reduction of pollutants did not increased significantly with the rise in temperature. In 1 h, the percentage reduction of COD was 79.21%, 81.62%, 83.65%, 84.71%, and phenol was 86.04%, 87.52%, 89.56%, 90.43%, respectively, at 30, 40, 50 and 60 °C. From these data we can assume that 50 °C is the optimum temperature. After 50 °C, the COD and phenol did not reduce significantly when T was increased to 60 °C.

Close

Fig. 5 Effect of temperature on (a) COD and (b) phenol reduction during Fenton oxidation process·H₂O₂ = 0.30 M, FeSO₄ = 3 g/L(0.043 M), COD_i = 2810.0 mg/L, Phenol_i = 283.0 mg/L.

Export to PPT

In a study the 40–50% COD reduction and 95% phenol reduction was achieved in 1 h, at operating condition pH < 6.5, H₂O₂ = 0.3 M and ferrous sulphate of 3 g/L at room temperature (Libing et al., 2012). Our results are better to this as 83.65% COD, 89.56% phenol and 79.34%, cyanide reduction were obtained in 1 h at 50 °C. Libing et al. (2012) has used the zero valent iron (Fe^o) powder at room temperature which was prepared from FeSO₄, in our studies also ferrous sulphate was used. In another studies electrocoagulation process has been also reported to treat CWW in which 91.18% COD, 42.85% phenol and 78.84% cyanide removal were achieved (Choudhary and Chaudhari, 2019).

3.6 Optimization using response surface methodology

3.6.1

3.6.1 Experimental design and optimization of fenton oxidation process

RSM is widely used for the designing of experiments by using mathematical and statically technique to determine the optimum conditions for desirable response (Draper and John, 1988). In the present study three independent variables for Fenton oxidation of CWW at three different levels based on its minimum to maximum range was chosen which are t_R, Cw and H₂O₂ concentration as shown in Table 2. The Central Composite Design (CCD) was performed by Design expert using MINITAB version 17.1.0.0.

Table 2 Experimental range for independent variables for Fenton oxidation process of CWW.

Independent Variables	Factors		Range and Levels
	Xi	−1	0	1
Reaction time (h)	X1	0	1	2
Ferrous salt (g/L)	X2	0.1	1.55	3
Hydrogen peroxide (M)	X3	0.07	0.22	0.36

Design of experiments was done to study the effect of t_R, Cw and H₂O₂ concentration on the COD and phenol reduction efficiency. In the present work only two way interactions were considered. Regression model was used to validate the second order polynomial to the experimental data and to predict the related model. The actual and predicated responses (Y) of COD and phenol reduction of total 20 runs were generated, and the removal efficiency of each experiment was calculated by using Eqs. (9) and (10) for percentage removal of COD and phenol. The experimental values and predicated values from Eqs. (9) and (10) are listed in Table 3 and shown in Fig. 6a, b. Quadratic models Eqs. (9) and (10) were fitted to the experimental data to obtain the regression model between the response (Y) and the independent variables (X1, X2, X3). The results of ANOVA are presented in Table 4.

(9)

$$\mathrm{Y}(\%\mathrm{C}\mathrm{O}\mathrm{D}\mathrm{r}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{l})=1.09+64.76\mathrm{X}1+8.334\mathrm{X}2+3.724\mathrm{X}3-21.938\mathrm{X}1\mathrm{X}1-0.723\mathrm{X}2\mathrm{X}2-0.1528\mathrm{X}3\mathrm{X}3+0.157\mathrm{X}1\mathrm{X}2+0.3440\mathrm{X}1\mathrm{X}3-0.2548\mathrm{X}2\mathrm{X}3$$

(10)

$$\mathrm{Y}(\%\mathrm{p}\mathrm{h}\mathrm{e}\mathrm{n}\mathrm{o}\mathrm{l}\mathrm{r}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{l})=-14.92+92.51\mathrm{X}1+8.01\mathrm{X}2+4.68\mathrm{X}3-36.32\mathrm{X}1\mathrm{X}1-1.95\mathrm{X}2\mathrm{X}2-0.2805\mathrm{X}3\mathrm{X}3+2.325\mathrm{X}1\mathrm{X}2+1.502\mathrm{X}1\mathrm{X}3-0.140\mathrm{X}2\mathrm{X}3$$

Table 3 RSM optimization of COD and phenol reduction for oxidation of CWW using Fenton process.

Run Order	X1	X2	X3	%COD Observed	%COD Predicated	%Phenol Observed	%phenol Predicated
1	1	0.10	7.5	67.140	66.47	68.080	72.76
2	1	3.00	7.5	77.980	79.06	83.590	82.17
3	1	1.55	7.5	74.430	74.29	82.650	81.56
4	2	1.55	7.5	75.714	76.05	83.240	80.00
5	1	1.55	7.5	74.430	74.29	82.650	81.56
6	2	0.10	2.5	53.670	53.72	42.830	43.50
7	0	0.10	12.5	23.420	24.27	0.520	0.34
8	0	3.00	2.5	26.350	26.03	0.706	0.47
9	1	1.55	7.5	74.430	74.29	82.650	81.56
10	0	0.10	2.5	10.670	10.21	0.360	0.44
11	2	3.00	2.5	71.420	70.45	62.320	61.69
12	2	0.10	12.5	74.460	74.67	78.690	78.11
13	1	1.55	12.5	78.640	77.37	84.860	83.33
14	2	3.00	12.5	83.660	84.01	88.460	92.24
15	1	1.55	7.5	74.430	74.29	82.650	81.56
16	1	1.55	7.5	74.430	74.29	82.650	81.56
17	1	1.55	7.5	74.430	74.29	82.650	81.56
18	0	3.00	12.5	32.870	32.71	2.460	0.97
19	1	1.55	2.5	61.880	63.56	60.980	65.77
20	0	1.55	7.5	28.570	28.64	3.990	10.50

Close

Fig. 6 The actual and predicted percentage reduction (a) COD and (b) phenol.

Export to PPT

Table 4 ANOVA regression model for (a) COD (b) phenol removal using Fenton process.

(a)
Source	Coefficient estimate	Sum of squares	Degree of freedom	Mean square	F value	P value	Remark
Model		9771.59	9	1085.73	1283.87	<0.0001	significant
Intercept	74.29
X1	23.70	5618.80	1	5618.80	6644.20	<0.0001
X2	6.29	395.89	1	395.89	468.14	<0.0001
X3	6.91	476.93	1	476.93	563.97	<0.0001
X1X2	0.2275	0.4140	1	0.4140	0.4896	0.5001
X1X3	1.72	23.67	1	23.67	27.99	0.0004
X2X3	−1.85	27.31	1	27.31	32.29	0.0002
X12	−21.94	1323.64	1	1323.64	1565.20	<0.0001
X22	−1.52	6.35	1	6.35	7.50	0.0209
X32	−3.82	40.11	1	40.11	47.43	<0.0001
Residual		8.46	10	0.8457
Lack of Fit		8.46	5	1.69
Pure Error		0.0000	5	0.0000

(b)
Source	Coefficient estimate	Sum of squares	Degree of freedom	Mean square	F value	P value	Remark
Model		23136.5	9	2570.72	173.7	<0.0001	significant
Intercept	81.56
X1	34.75	12075.9	1	12075.9	815.96	<0.0001
X2	4.71	221.43	1	221.43	14.96	0.0031
X3	8.78	770.78	1	770.78	52.08	<0.0001
X1X2	3.37	90.95	1	90.95	6.15	0.0326
X1X3	7.51	451.29	1	451.29	30.49	0.0003
X2X3	−1.02	8.25	1	8.25	0.5577	0.4724
X12	−36.32	3627.13	1	3627.13	245.08	<0.0001
X22	−4.1	46.17	1	46.17	3.12	0.1078
X32	−7.01	135.23	1	135.23	9.14	0.0128
Residual		148	10	14.8
Lack of Fit		148	5	29.6
Pure Error		0	5	0

3.6.3

3.6.3 Interaction effects of independent variables

The interaction effect of independent parameters like H₂O₂ concentration, Cw and t_R can be understand from Table 4a and b and shown in Fig. 7. The experimental results indicates that interaction of Cw and t_R has less effect on COD reduction (F = 0.49 and p = 0.50), while Cw and H₂O₂ interaction has less effect on phenol reduction (F = 0.56 and p = 0.47). When H₂O₂ concentration was increased with time, the percentage removal of COD and phenol was also increased. The removal efficiencies became significant with p value 0.000 for COD and 0.000 for phenol removal as presented in Table 4a and b and shown in Fig. 7. Similarly, when the Cw was increased with H₂O₂ concentration, the COD percentage removal was increased with having p = 0.0002, but phenol removal was not increased much and has p = 0.472 as shown in Fig. 7 and Table 4.

Close

Fig. 7 Three dimensional (3D) response surface graphs of COD reduction (a, b, c) and phenol reduction (d, e, f) for Fenton process.

Export to PPT

CWW degradation using Fenton process is considered as free radical mechanism. With proper maintaining the parameters like H₂O₂ concentration and Cw in reactor, pollutants can be degraded easily. Although, Fenton oxidation process had a good percentage reduction of pollutants contained in CWW, even for the complete reduction of pollutants, some effective and commercial process need to apply as second stage treatment. There are many processes, but the adsorption was our choice, because it is an effectiveness and low cost process (Rafatullah et al., 2010; Burmistrz et al., 2014).

From ANOVA the optimum condition of various parameters obtained after examining the response curves and contour plots was Cw = 1.85 g/L (0.0266 M), H₂O₂ = 0.33 M and t_R = 1.52 h. At this operating condition maximum 84.66% COD and phenol reduction was obtained in Fenton oxidation process.

3.7 Adsorption process to treat pre-treated CWW using Fenton process

Adsorption is a surface phenomenon in which high surface adsorbent is used to remove solute contained in air or liquid for the removal of pollutants (Rafatullah et al., 2010; Burmistrz et al., 2014). Among various adsorbent activated carbon is the preferred adsorbent because of its high efficiency and application in the treatment of large scales. In the present study CWW was first treated by Fenton oxidation process then treated by adsorption process as second stage treatment using activated carbon as adsorbent. Effect of pH = 3, 4.5, 6, 7.5, 9, and adsorbent dose Cw = 1, 2, 3, 4, 5 g/L were studied on the removal of pollutants having initial concentration of COD_i = 590.0 mg/L, phenol = 39.49 mg/L and cyanide = 5.2 mg/L.

3.7.1

3.7.1 Effect of pH

The solution pH has been found to significant effect on adsorption process as pH alters the charges of adsorbent and species contained in wastewater. Effect of pH was studied in the range of 3–9 on the removal of COD and phenol are shown in Fig. 8a, b. Initial concentration of COD (COD_i) was 590.0 mg/L and phenol (phenol_i) was 39.49 mg/L COD reduction of 68.83%, 57.77%, 62.76%, 64.10% and 50.32% obtained, while phenol reduced to 99.21%, 85.66%, 95.04%, 90.32% and 82.65% at pH 3, 4.5, 6, 7.5 and 9, respectively, in 2 h with 4 g/L adsorbent. From the findings we can say that the pH 3 is the best because it gave the highest removal. Since, Fenton oxidation performs best at pH = 3, therefore no pH adjustment was required for adsorption process. The data shows phenol removal was little affected by pH solution, which is due to neutral charge of phenol, while, COD reductions were affected with pH because CWW contain various species, including phenol. The cyanide, thiocyanate, ammonium, sulphate, phenolic compounds, polynuclear aromatic hydrocarbons and polycyclic nitrogen-containing acyclic compounds are present in CWW. These compounds have different functional groups which has different charge. A part from this formation of some other compounds is expected during the Fenton oxidation of CWW.

Close

Fig. 8 Effect of pH on (a) COD and (b) phenol removal by adsorption process. Adsorbent dosage = 4 g/L, T = 30 °C, t_R = 3 h, COD_i = 590.0 mg/L, Phenol_i = 39.49 mg/L.

Export to PPT

3.7.2

3.7.2 Effect of adsorbent dosage

Adsorbent effect was studied by varying activated carbon dosage from 1 to 5 g/L. The results are presented in Fig. 9a, b. As the adsorbent dosage was increased, reduction of pollutants was also increased with Cw = 1, 2, 3, 4 and 5 g/L, respectively, the 22.16, 48.33, 54.16, 68.83 and 69.00% COD reduced from COD_i = 590.0 mg/L, and 85.97, 91.92, 96.80, 99.21 and 100.00% phenol removal obtained from phenol_i = 39.49 mg/L.The reason for increase in removal with increase in adsorbent mass loading is availability of more active sites for the adsorption of phenol. The rate of adsorption was fast at first 1 h, after that it was slow. After 2 h there was no significant reduction in pollutant was noted.

Close

Fig. 9 Effect of adsorbent dosage on (a) COD (b) phenol removal. T = 30 °C, t_R = 3 h, COD_i = 590 mg/L, Phenol_i = 39.49 mg/L.

Export to PPT

3.8 Analysis of CWW through SEM and EDX

CWW, Fenton treated sludge obtained after Fenton process and adsorption treated CWW was characterised by scanning electron microscopy (SEM) and energy dispersive X ray (EDX) techniques to observe morphology and to identify the major elements present before and after treatment. Fig. 10 shows the SEM images of untreated, treated by Fenton oxidation, sludge obtained from Fenton oxidation, and treated effluent by Fenton oxidation and adsorption. The different nature of images can be seen. The EDX figure shows various elements present in CWW before and after treatment. The various elements contained are presented in Table 5. In the table in can be seen that various elements amount were reduced after treatment. Less amount of carbon in treated CWW is due to its reduction during treatment process, the low COD of treated CWW verify to this.

Close

Fig. 10 EDX of untreated CWW (a) untreated CWW (b) Fenton treated CWW (c) sludge obtained from Fenton oxidation (d) Fenton oxidation and adsorption treated CWW.

Export to PPT