A systematic analysis was carried out by using response surface methodology to create a quantitative model of the synergistic effects of conditions in a continuous freezer [mix flow rate (L/h), overrun (%), cylinder pressure (kPa), drawing temperature ( degrees C), and dasher speed (rpm)] on the principal constituent parameters of ice cream [rate of fat destabilization (%), mean air cell diameter (mum), and mean ice crystal diameter (mum)]. A central composite face-centered design was used for this study. Thirty-one combinations of the 5 above-mentioned freezer conditions were designed (including replicates at the center point), and ice cream samples were manufactured and examined in a continuous freezer under the selected conditions. The responses were the 3 variables given above. A quadratic model was constructed, with the freezer conditions as the independent variables and the ice cream characteristics as the dependent variables. The coefficients of determination (R(2)) were greater than 0.9 for all 3 responses, but Q(2), the index used here for the capability of the model for predicting future observed values of the responses, was negative for both the mean ice crystal diameter and the mean air cell diameter. Therefore, pruned models were constructed by removing terms that had contributed little to the prediction in the original model and by refitting the regression model. It was demonstrated that these pruned models provided good fits to the data in terms of R(2), Q(2), and ANOVA. The effects of freezer conditions were expressed quantitatively in terms of the 3 responses. The drawing temperature ( degrees C) was found to have a greater effect on ice cream characteristics than any of the other factors.

The effect of conventional continuous freezer parameters [mix flow (L/h), overrun (%), drawing temperature ( degrees C), cylinder pressure (kPa), and dasher speed (rpm)] on the hardness of ice cream under varying measured temperatures (-5, -10, and -15 degrees C) was investigated systematically using response surface methodology (central composite face-centered design), and the relationships were expressed as statistical models. The range (maximum and minimum values) of each freezer parameter was set according to the actual capability of the conventional freezer and applicability to the manufacturing process. Hardness was measured using a penetrometer. These models showed that overrun and drawing temperature had significant effects on hardness. The models can be used to optimize freezer conditions to make ice cream of the least possible hardness under the highest overrun (120%) and a drawing temperature of approximately -5.5 degrees C (slightly warmer than the lowest drawing temperature of -6.5 degrees C) within the range of this study. With reference to the structural elements of the ice cream, we suggest that the volume of overrun and ice crystal content, ice crystal size, and fat globule destabilization affect the hardness of ice cream. In addition, the combination of a simple instrumental parameter and response surface methodology allows us to show the relation between freezer conditions and one of the most important properties-hardness-visually and quantitatively on the practical level.

The primary process parameters of the homogenization pressure and freezing process (drawing temperature and overrun) for ice cream manufacture were examined to determine their impact on the sensory attributes and odor sensor response of ice cream. Fifteen process conditions were selected using a Box-Behnken design, while 12 sensory attributes were obtained as assessment items based on sensory evaluations using quantitative descriptive analysis (QDA). Eleven of these sensory attributes changed significantly according to process conditions, suggesting that such conditions can have a major impact on ice cream's sensory attributes, even for a fixed make-up of ingredients. Furthermore, observed correlations between the sensory attributes and microstructural attributes of the ice cream led to the conjecture that the sensory attributes were influenced by changes to the ice cream's structural conditions resulting from the process conditions. A correlation was also observed between the odor sensor response and the overrun condition, but no clear correlations were found to exist within the ice cream structure or the sensory attributes. * Glyceryl mono-distearate ** 45.0% locust bean gum, 45.0% guar gum, and 10.0% carrageenan