In the first case, the calculation was carried out using a non linear
least-squares (quasi-Newton) method via the macro ‘Solver’ in
the Microsoft Excel XP spreadsheet. Later, the Statistica 6.0 program
(StatSoft, Inc. 2001) was used to calculate the significance
of the estimated parameters (Student t-test, a = 0.05) and the
robustness of the model (Fisher F-test, a = 0.05). Results of the factorial
designs were employed to obtain empirical equations that
describe the significant parameters as a function of temperature
and ethanol concentration. The statistical significance of the coefficients
was verified by means of the Student t-test (a = 0.05),
and the model consistency by the Fisher F-test (a = 0.05) using
the following mean squares ratios: