Monitoring flavor development during the roasting of cereals

Roasted cereal products are important sources of flavor and color in the brewing industry. However, production to a flavor specification, although desired by customers, is not yet possible. Samples of commercial roasted products (amber, chocolate, black malt, and roasted barley) were sampled periodically during roasting. Two techniques were then used to follow the development of flavor: analytically using chemical marker compounds representing the major groups of flavor-active compounds found in roasted cereals, and flavor profiling by an expert taste panel. Principal components analysis was used to investigate relationships between the chemical and sensory data. The majority of the measured chemical marker compounds were not detected until after an hour of roasting and, in the case of amber malt, certain thiazoles were not detected at all. The formation of these compounds is a function of moisture, temperature, or both. Flavor development preceded color development and, in all cases, the greatest development of color occurred in the last 10-15 min of roasting. Sensory analysis showed that high-color end products were less bitter, burnt, and harsh than were samples taken earlier in the roasting time course. This was corroborated by gas chromatographic data, which showed a reduction in the concentration of sulfur and nitrogen heterocyclic compounds (particularly pyrazines) toward the end of roasting. None of the sulfur or nitrogen compounds detected in amber malt decreased in concentration toward the end of roasting. This again appears related to moisture level and product temperature. It may be feasible, therefore, to promote or inhibit the formation of certain flavors just by manipulating the air temperature/flow entering the roaster. This will affect moisture level and product temperature. which, in turn, will affect flavor compound formation.