Sweetener Blend Optimization by Using Mixture Design Methodology and the Electronic Tongue

Abstract

Utilizing more than one sweetener has been shown to be an effective way to substitute sucrose in food products. The objective of this study was to apply the augmented simplex-centroid mixture design for the optimization of acceptable sweetener blends using coconut sugar, agave, and stevia. Sweetener blends were evaluated in aqueous solutions and gluten-free granola bars by a trained panel and consumers (n = 60). Significant differences were found between sweetener mixtures in solutions by both panelists and consumers (P < 0.05). Taste profiles for the sweetener solutions were also generated using the electronic tongue. Most consumer and trained intensity ratings were highly correlated (R2 ≥ 0.79) with the electronic tongue taste profile analysis. Granola bars were also found to be significantly different (P < 0.05), with consumers preferring coconut sugar mixtures. Using contour plots and desirability function analysis, an optimal sweetener combination was found for a granola bar formulation of 89.9% coconut sugar, 6.1% agave, and 4% stevia. These results indicate that a mixture design can be a reliable way to develop new sweetener blends for product development. Practical Application: This study demonstrates how blending sweeteners using a mixture design can create a more accepted flavor profile and allows more flexibility in sweetener choice for product developers based on production and quality constraints. The results also indicate the potential of coconut sugar to be used in commercial food and beverage products. In addition, the electronic tongue was shown to have the capacity to predict sweetener differences, an important ability when determining sweetener substitutes in food and beverage products.