The experimental design was statistically analysed using the
Design Expert 7.1.1 software (Stat-Ease, Inc., Minneapolis, MN).
The set of experiments followed the BoxeBehnken design (Box &
Behnken, 1960), formed by combining two-level factorial designs
with incomplete block designs. This procedure creates designs with
desirable statistical properties but with only a fraction of the experiments
required for a three-level factorial design. Error assessment
was based on a replication of the central point for each
storage time (0, 1, and 3 months) as suggested in the BoxeBehnken
design. The following second order polynomial modelwas used as a
first approach to analyse the experimental data: