Optimization of additive combination for improved soy-wheat bread quality

PD RIBOTTA; P ROCCIA; GT PEREZ; AE LEON; MC AÑON

Rosario, Argentina

Congreso; 1a ICC Latinoamerican Conference on Cereals and Cereal Products; 2007

ICC

Blending of 10% soy flour with wheat flour not only provides higher protein content but also improves the amino acid balance. However, the incorporation of high levels of soy products have negative effect in gluten network formation, extensibility properties and gas retention of dough and final bread quality. The combination of additives could help to overcome these problems. The present study was designed: to determine the influence and interactions of sodium stearoyl-2-lactylate, transglutaminase and hemicellulase on high-protein bread quality; and to improve bread quality of soy-wheat bread by using an optimized additive combination. Wheat flour was blended with heat-treated defatted soy flour (90:10 respectively). Tested additives: Sodium stearoyl-2-lactylate (SSL), transglutaminase (TG, activity: 100 U/g) and hemicellulase (HE, activity: 2500 FXU(W)/g). Specific loaf volume (SLV), crust color and crumb texture were determined. Besides, crumb-structure characteristics were measured from digitalized images: cell to total area ratio and cell uniformity (the ratio of the number of cells with areas less than, to those greater than, 4 mm2). A central composite design was used to arrange experiments and four replicates were made at the centre point. Independent variables: SSL (0.10 and 0.50 g/100 g flour), HE (0.005 and 0.02 g/100 g flour) and TG (0.05 and 0.30 g/100 g flour). Levels of variables were selected on the basis of manufacturer recommended levels. The results were analyzed by a multiple regression method. Quality of the models fitness was evaluated by ANOVA (Statgraphics plus 5.0). The experimental results were applied to obtain the regression models. Based on experimental design the cubic model could be fitted in case the lacks of fit of the quadratic or linear models were statistically significant (p0.05) with an elevated determination coefficient (0.986, 0.967 and 0.970, respectively). For the bread crumb structure, only the cubic model obtained from cell uniformity values presented an elevated determination coefficient (0.946) although it had a significant lack of fit (p=0.033). For the color parameters, it was not possible to fit a model for L*, a* and b*. The result of the optimization process suggested a mixture containing 0.540% of SSL, 0.009% of HE and 0.068% of TG as the best mixture of these three additives in order to achieve the maximum specific bread volume and the minimum hardness and chewiness. From the data presented, it was evident that soy-wheat bread could be baked and that good quality (high SLV and cell uniformity and low crumb hardness) could be obtained by optimizing additive mixture. These breads with elevated levels of protein (~15%) and fibers can be produced on an industrial scale.