Cr VI catalytic determination

In a full factorial design, the runs at all possible combinations of the selected factor levels should be performed in random order, as a preventive measure against bias in the response values. Sometimes this 

The linear model shows lack of fit. The quadratic and special cubic models are represented by the expressions 

Design for the catalytic determination of Cr(VI). xi, X2 and X3 are the pseudocomponent proportions of water, acetone and N,N-dimethylformamide 

Trying to fit the usual models to the data in the table, we find that this system is more complex than the others we have seen in this chapter. All of the simpler models presented lack of fit, and only the full cubic model proved satisfactory for both responses. In terms of the original quantities, these models are given by the equations below, where only the significant terms (at the 95% confidence level) are retained. 

Log(cond) = -9ci -h 327c2 - 601ciC2 - 920c2C3 -h 1090ciC2C3 - -334ciC2(ci - C2) 

---