Abstract
The Maillard reaction in foods has been studied by using a computer simulation technique. As a typical model system, the dehydration process of an ordinary leek was effectively used to demonstrate an optimum drying operation. The Maillard reaction rate of the leek has been measured as a color change using a commercially prepared color meter. To evaluate the color change of the leek, a color parameter of ?E* (=((?a*)2+(?b*)2+(?L*)2)1/2) was chosen because it followed a good linear relation with the dehydration response curve of the leek. The response curves of ?E* obtained in the dehydration operation were described by a typical consecutive reaction model A?B?C, where A is glucose-like materials contained in the leek, B is intermediates formed, and C is melanoidin-like materials as final products. The computer simulation fitting to both the experimental dehydration response curves of CC and the ?E*-value response curves evaluated the maximum amount of CB (CBmax).
The integrated amount of CB (CBin) produced in the course of the dehydration process changed appreciably depending on the drying temperature and the humidity of the drying air. This took place because the water species dynamically shifted from species A1 (weakly restricted water species in the A1 region of W0 = 1600~120 %-d.b.) to A2 (strongly restricted water species in the A2 region of W0 = 120~20 %-d.b.) at the water content of W0 = 120 %-d.b.
The sensory scores, evaluated by both the four human tasters and a commercially distributed taste tester as a function of CBin, gave a gradual increase curve indicating the larger CBin to be the higher taste score. A two-step dehydration, using the 70 °C - operation in the water species A1 region and the 40 °C - operation in the water species A2 region, was recognized as an optimized operation for the best sensory score product.
