Table 1 shows the results of 13 experimental runs carried
out using the design protocol. The data obtained was analyzed
and fitted into various polynomial models. It was observed that
the response Y1 (particle size) fitted best into the quadratic
response surface model with backward elimination after natural logarithmic transformations of the data. On the other hand
response Y2 (% encapsulation efficiency) fitted best into the
quadratic response surface model after square root transformation
of the data.
The polynomial models showing the relationship between
the independent variables and the responses Y1 and Y2 can be
expressed by the following Eqs. (3) and (4):
Ln(Y1) = 6.47 + 0.40X1 + 0.29X2 + 0.252
1 − 0.28X2
2 (3)
Sqrt(Y2) = 3.31 − 1.11X1 + 0.23X2 + 0.42X1X2 + 0.20X2
1 − 0.17X2
2
(4)