3. Results
3.1. Response surface analysis
As shown in Table 1, the arrangement of CCD was in such a way that allows the development of the appropriate empirical equations. Randomized experiments were carried out in order to reduce the effects of unexplained variability in the actual responses caused by extraneous parameters. In order to assess the repeatability of the method, the center point was repeated six times. Interaction, quadratic, and linear (main) effects of independent factors were assessed for the response variables (Yi) by RSM. Table 2 presented two high-pressure variables regression coefficients, along with the corresponding R2, lack of fit test, p-values, F-ratios and their significant probability of each parameter. The final reduced models fitted for MPS and PDI demonstrated high coefficients of determination (0.995 and 0.965) and (0.953 and 0.965) for nano-sized calcium carbonate-enriched-milk and nano-sized calcium citrate-enriched-milk, respectively. To describe the variation of responses as the function of the main preparation parameters, the satisfactory adjustment of the reduced response surface models were employed. PDI and MPS of nano-sized calcium carbonate-enriched-milk and nano-sized calcium citrate-enriched-milk were significantly (p = 0.00) influenced by pressure and cycle (Table 2).