Table 2 Analysis of variance for the response surface of Cu2+ removal % by Azotobacter nigricans NEWG-1 obtained by the Box-Behnken design.

From: Statistical modeling-approach for optimization of Cu2+ biosorption by Azotobacter nigricans NEWG-1; characterization and application of immobilized cells for metal removal

Source of variance

Degrees of freedom

Contribution, %

Sum of square

Mean of square

F-value

P-value

Coefficient estimate

Overall model

9

98.7

6163.84

684.87

43.49

0.0003*

58.89

Linear effect

X1

1

0.7

45.41

45.41

2.88

0.1502

2.38

X2

1

64.6

4029.78

4029.78

255.92

<0.0001*

22.44

X3

1

12.9

807.02

807.02

51.25

0.0008*

10.04

Total linear

3

78.2

4882.2

1627.4

103.4

0.000*

 

Square effect

X12

1

5.1

340.93

340.93

21.65

0.0056*

−9.61

X22

1

8.7

494.41

494.41

31.40

0.0025*

−11.57

X32

1

2.6

159.30

159.30

10.12

0.0245*

6.57

Total square

3

16.4

1020.6

340.2

21.6

0.003*

 

Interaction effect

X1X2

1

0.3

17.35

17.35

1.10

0.3420

2.08

X1X3

1

1.4

87.70

87.70

5.57

0.0647

−4.68

X2X3

1

2.5

156.00

156.00

9.91

0.0254*

−6.25

Total interaction

3

4.2

261.1

87.0

5.5

0.048*

 

Error effect

Pure Error

2

0.1

6.85

3.42

   

Lack-of-Fit

3

1.2

71.88

23.96

7.0

0.1276

 

Total error

5

1.3

78.7

15.8

   

Overall total

14

100.0

     

R2

98.74%

Std. Dev.

3.97

     

Adj R2

96.47%

Mean

51.10

     

Pred R2

81.33%

C.V.%

7.77

     

Adeq Precision

20.1125

PRESS

1165.54

     
  1. *Indicates significant effect. “Std. Dev. is the standard deviation, the coefficient of determination (R2), Adj R2 is the adjusted-R2, Pred R2 is the predicted-R2 and PRESS is the prediction error sum of squares, Adeq Precision is adequate precision, C.V is the Coefficient of variation”.