The results of the analysis of variance test is summarized in Table 4. The probability > F for the model is less than 0.05 which implies that the model is significant and the terms in the model have significant effects on the response. In this case A, B, C, D, AB, AC, AD, BC, BD, CD, A2, B2, C2, D2 are significant model terms at the 95 % confidence level (α =5 %). The model F-value of 1387.59 and P-value of < 0.0001 implies that the model is highly significant. Based on the ANOVA results, the values of R2, Adjusted R2 and Predicted R2 were 0.9992, 0.9985 and 0.9971, respectively. This result suggests that the regression model is well interpreted the relationship between the independent variables and the response. Furthermore, the adequate precision ratio of 149.08 in the study shows that this model could be applied to navigate the design space defined by the CCD.
Table 4 ANOVA results for Response Surface Quadratic Model
Source Sum of squares df Mean square F -Value  P- value
model 12060.48 1 861.46 1387.59 <0.0001
A-pH 914.15 1 914.15 1472.45 <0.0001
B-TCcon. 1730.94 1 1730.94 2788.08 <0.0001
C-PScon. 3999 1 3999 6441.32 <0.0001
D-Time 4676.04 1 4676.04 7531.85 <0.0001
AB 18.66 1 18.66 30.06 <0.0001
AC 30.97 1 30.97 49.88 <0.0001
AD 6.84 1 6.84 11.01 0.0047
BC 16.4 1 16.4 26.42 0.0001
BD 17.64 1 17.64 28.41 <0.0001
CD 16.61 1 16.61 26.75 0.0001
A2 571.64 1 571.64 920.76 <0.0001
B2 123.3 1 123.3 198.6 <0.0001
C2 29.96 1 29.96 48.26 <0.0001
D2 20.18 1 20.18 32.5 <0.0001
Residual 9.31 15 0.62
Lack of Fit 5.13 10 0.51 0.61 0.7607
Pure Error 4.18 5 0.84
Cor Total 12069.8 29
R2 = 0.9992; Adjusted R2 = 0.9985 and Predicted R2 = 0.9971