Supplemental materials

Design and optimization of a combined electrocoagulation-electroflotation reactor for wastewater treatment

C. Jiménez, C. Sáez, P. Cañizares, M.A. Rodrigoa

aDepartment of Chemical Engineering. Universidad de Castilla La Mancha, Campus Universitario 12. E13071 Ciudad Real. Spain; Fax: +34 926 29 52 56; Tel: +34 926 29 53 00; E-mail:

Experimental design

In this work, the effect of three operational inputs (current density (x1), residence time of the reactor (x2) and pollutant concentration (x3)) has been studied using a 23 experimental design combined with a star design using four replicates of the central point. For the star design, half of the range of the experimental design has been taken. So, the experimental design and the star design experiments can be shown in Table SM1 for kaolin, Table SM2 for EBT and Table SM3 for oil-in-water emulsions. According to literature(Box et al. 2002), all the experiments were randomly made.

Table SM1.Original and coded factors for kaolin suspensions treatment

Factor / Original factor / Coded factor
-1 / -0.5 / 0 / +0.5 / +1
Residence time/ h / X1 / 0.2 / 0.225 / 0.25 / 0.275 / 0.3
Current density/ mAcm-2 / X2 / 1 / 1.375 / 1.75 / 2.125 / 2.5
Kaolin concentration/ mg/L / X3 / 500 / 750 / 1000 / 1250 / 1500

Table SM2. Original and coded factors for EBT solution treatment

Factor / Original Factor / Coded Factor
-1 / -0.5 / 0 / +0.5 / +1
Residence Time/ h / X1 / 0.167 / 0.221 / 0.275 / 0.329 / 0.383
Current Density/ mAcm-2 / X2 / 2.5 / 3.7 / 5 / 6.2 / 7.5
EBT concentration/ mg/L / X3 / 50 / 100 / 150 / 200 / 250

Table SM3. Original and coded factors for oil-in-water emulsion treatment

Factor / Original Factor / Coded Factor
-1 / -0.5 / 0 / +0.5 / +1
Residence Time/ h / X1 / 0.15 / 0.225 / 0.30 / 0.375 / 0.45
Current Density/ mAcm-2 / X2 / 25.0 / 31.2 / 37.5 / 43.7 / 50.0
Oilconcentration/ mg/L / X3 / 1000 / 2000 / 3000 / 4000 / 5000

Experimental data were fitted to a second order polynomial model if the results lead to a non-linear system or to a first order polynomial model if they lead to a linear system, and the regression coefficients a0, ai, aii and aij were obtained in each case (equations sm1 and sm2).

(sm1)

(sm2)

where, a0, ai, aii and aij are the regression coefficients for intercept, linear, quadratic and interaction terms, respectively and xi, xj are the independent variables. Yi represents pollutant removal (Y1) and floated/settled solids ratio (Y2). Y1 indicates the efficiency of the process for pollutants removal, and Y2 represents the efficiency of the flotation process using gas bubbles generated in the cathode, related to sedimentation inside the reactor. Both parameters are of great importance as the main objective of this work is to remove pollutants from wastewater but trying to optimize the electroflotation process inside the reactor.

Central experiments

First, four series of experiments with the central conditions of the experimental design were made to calculate the experimental error (according to equation sm3) and the standard deviation (eq sm3) where e is the experimental error, t is the t-Student function value for a confidence level of 99% (taking into account 3 degrees of freedom this function gets a value of 5.841), s is the standard deviation and r the number of replications.

(sm3)

The experimental error has been calculated for both kaolin, EBT and emulsions assays and taking into account pollutant removal and the floated-settled solids ratio. The results for the four replications of the central conditions assays and the calculated experimental error are shown in Table SM4.

Table SM4. Experimental error obtained from central conditions of the experimental design. Confidence interval for calculate t-Studentvalue: 99%.

Pollutant / Central conditions / Experimental error
1 / 2 / 3 / 4
Kaolin / Turbidity removal/ % / 76.50 / 67.60 / 69.90 / 68.35 / 7.46
0.10
TFS/TSS* / 0.68 / 0.88 / 0.86 / 0.92
EBT / COD removal/ % / 52.69 / 56.41 / 55.88 / 49.95 / 8.79
Colour removal/ % / 64.34 / 60.23 / 71.84 / 61.82 / 15.02
VFS/VSS* / 14.34 / 12.64 / 13.25 / 13.52 / 2.06
O/W Emulsions / Turbidity removal/ % / 95.16 / 99.01 / 97.15 / 97.61 / 4.66
COD removal/ % / 89.64 / 83.30 / 94.21 / 84.60 / 14.59
TFS/TSS* / 12.31 / 10.68 / 12.17 / 10.62 / 2.69

*TFS=Total Floated Solids, TSS=Total Settled Solids, VFS=Volatile Floated Solids, VSS=Volatile Settled Solids

As it can be seen, the higher values of the experimental error are obtained for colour removal in the case of working with EBT solutions, and for COD removal in the case of oil-in-water emulsions. Taking into account these results, the parameters selected for the study have been COD for EBT solutions and turbidity removal for kaolin suspensions and oil-in-water emulsions.

Experimental design

The experimental design has been made for two main parameters: pollutant removal (COD has been considered in the case of EBT solutions and turbidity for oil-in-water emulsions and kaolin suspensions) and floated-settled solids ratio. The experimental conditions and the results obtained for both parameters in steady state conditions are shown in Table SM5 for kaolin suspensions, in Table SM6 for EBT solutions and in Table SM7 for oil-in-water emulsions. Although the experiments are shown in order in the tables, although they were randomly made.

Table SM5. Experimental results obtained for the experimental design for Kaolin suspensions. Pollutant: Kaolin; Temperature: 25 ºC; electrolyte: 5000 mg/LNaCl.

Assay / Residence time/ h / Current density/ mAcm-2 / Kaolin concentration/
mg/L / Turbidity removal/ % / TFS/TSS
1 / 0.300 / 1.0 / 500 / 31.87 / 0.98
2 / 0.200 / 1.0 / 500 / 13.48 / 1.03
3 / 0.300 / 2.5 / 500 / 62.39 / 0.99
4 / 0.200 / 2.5 / 500 / 33.21 / 0.95
5 / 0.300 / 1.0 / 1500 / 55.09 / 0.77
6 / 0.200 / 1.0 / 1500 / 49.50 / 0.45
7 / 0.300 / 2.5 / 1500 / 55.30 / 0.55
8 / 0.200 / 2.5 / 1500 / 49.86 / 0.64
9 / 0.250 / 1.4 / 1000 / 34.79 / 0.94
10 / 0.250 / 2.1 / 1000 / 51.93 / 0.68
11 / 0.275 / 1.7 / 1000 / 50.22 / 0.56
12 / 0.225 / 1.7 / 1000 / 41.59 / 0.53
13 / 0.250 / 1.7 / 1250 / 53.18 / 0.48
14 / 0.250 / 1.7 / 750 / 39.64 / 0.84

Table SM6. Experimental results obtained for the experimental design for EBT solutions. Pollutant: EBT; Temperature: 25 ºC; electrolyte: 5000 mg/LNaCl.

Assay / Residence time/ h / Current density/ mAcm-2 / EBT concentration/
mg/L / COD removal/ % / VFS/VSS
1 / 0.167 / 2.5 / 50 / 25.24 / 9.74
2 / 0.383 / 2.5 / 50 / 40.00 / 15.84
3 / 0.167 / 7.5 / 50 / 42.86 / 4.27
4 / 0.383 / 7.5 / 50 / 43.14 / 6.99
5 / 0.167 / 2.5 / 250 / 7.14 / 10.03
6 / 0.383 / 2.5 / 250 / 18.61 / 5.44
7 / 0.167 / 7.5 / 250 / 40.29 / 13.12
8 / 0.383 / 7.5 / 250 / 67.23 / 12.77
9 / 0.275 / 3.7 / 150 / 5.45 / 4.16
10 / 0.275 / 6.3 / 150 / 70.75 / 12.22
11 / 0.221 / 5.0 / 150 / 18.63 / 18.24
12 / 0.329 / 5.0 / 150 / 69.98 / 10.49
13 / 0.275 / 5.0 / 100 / 57.22 / 16.25
14 / 0.275 / 5.0 / 200 / 38.66 / 10.13

Table SM7. Experimental results obtained for the experimental design for oil-in-water emulsions. Pollutant: O/W emulsions; Temperature: 25 ºC; electrolyte: 10000 mg/LNaCl.

Assay / Residence time/ h / Current density/ mAcm-2 / Oil concentration/ mg/L / COD removal/ % / TFS/TSS
1 / 0.150 / 25.0 / 1000 / 97.92 / 8.91
2 / 0.450 / 25.0 / 1000 / 99.67 / 1.96
3 / 0.150 / 50.0 / 1000 / 99.31 / 6.71
4 / 0.450 / 50.0 / 1000 / 99.42 / 3.30
5 / 0.450 / 50.0 / 5000 / 99.66 / 3.38
6 / 0.150 / 50.0 / 5000 / 92.69 / 9.62
7 / 0.450 / 25.0 / 5000 / 97.93 / 4.22
8 / 0.150 / 25.0 / 5000 / 76.12 / 5.26
9 / 0.225 / 37.5 / 3000 / 99.29 / 6.05
10 / 0.375 / 37.5 / 3000 / 99.54 / 3.90
11 / 0.300 / 31.2 / 3000 / 98.95 / 3.35
12 / 0.300 / 43.7 / 3000 / 99.32 / 3.63
13 / 0.300 / 37.5 / 2000 / 99.61 / 6.10
14 / 0.300 / 37.5 / 4000 / 99.14 / 6.39

According to the response surface methodology, these results have been fitted to a second order polynomial model and the regression coefficients have been calculated according to equation sm4. The regression coefficients for the model in each case are shown in Table SM8. The experimental results are compared with the results obtained with the model in Figure SM1.

Y= a+b·J+c·θ+d·C+e·θJ+f·JC+g·θC+h·J2+i·θ2+j·C2+k·θJC(sm4)

Where J is the current density, θ the resident time and C the pollutant concentration.

Table SM8. Regression coefficients for the second order polynomial model for turbidity removal and TFS/TSS ratio (kaolin suspensions), COD removal and VFS/VSS ratio (EBT solutions) and turbidity removal and TFS/TSS ratio (O/W emulsions)

Parameter / Kaolin / EBT / O/W emulsions
Turbidity removal/ % / TFS/TSS / COD removal/ % / VFS/VSS / Turbidity removal/ % / TFS/TSS
a / -59.845 / -0.755 / -66.413 / -3.393 / 107.748 / 6.750
b / 65.560 / 0.166 / 26.782 / 9.858 / -0.116 / 0.175
c / -401.447 / 1.400 / 300.356 / -19.570 / -15.283 / 0.500
d / 0.124 / 8.100·10-4 / -0.160 / 1.500·10-3 / -0.013 / -2.4·10-3
e / 0.000 / -0.465 / -18.740 / -3.769 / 0.189 / -0.414
f / -0.019 / -6.030·10-4 / -8.000·10-4 / 8.600·10-3 / 2.126·10-4 / 3.000·10-5
g / -0.192 / -4.170·10-3 / -0.336 / -0.402 / 0.027 / 8.000·10-4
h / -10.368 / - / -261.095 / -1.137 / - / -1.300·10-3
i / 1494.000 / - / 0.000 / 117.872 / - / 0.570
j / -1.158·10-5 / - / 4.671·10-5 / 4.480·10-5 / - / 1.980·10-7
k / 0.000 / 2.000·10-3 / 0.143 / 0.037 / -4.277·10-4 / 1.690·10-5

Values of adjusted r2 of 0.935, 0.7944 and 0.9999 are obtained for pollutant removal for kaolin, EBT and oil-in-water emulsionsand values of 0.927, 0.661 and 0.734 are obtained for floated-settled solids ratio (for kaolin and emulsions, total solids have been measured, as for EBT solutions volatile solids have been considered, as pollutant concentration is much lower than electrolyte concentration) (Figure sm1). Additionally, Response Surfaces have been obtained from these models for both pollutant removal and floated/settled solids ratio.

Fig. SM1Comparison of the modelled data with the experimental results. Left figures: Pollutant removal; right figures: Floated/settled solids ratio; a) kaolin suspensions; b) EBT solutions; c) oil-in-water emulsions. (TFS/TSS=Total Floated/Settled Solids; VFS/VSS=Volatile Floated/Settled Solids)

References

Box GEP, Hunter WG, Hunter JS, de Pablos LA, Llabres JTM (2002): Estadística para investigadores: introducción al diseño de experimentos, análisis de datos y construcción de modelos. Reverté