Effects of time and temperature during melanging on the volatile profile of dark chocolate
Date
2019
Authors
Clark, Caitlin, author
Stone, Martha, advisor
Miller, Jeffrey, committee member
DiVerdi, Joseph, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
Small-batch chocolate-makers use distinct equipment and process steps from that of large-scale industry leaders. Only one of these steps is unique to small-batch chocolate-making and is known as melanging—a refining process that takes place in a stone wet-grinder. Despite being essential to flavor and overall quality development, it has never been studied, and all data on its application to chocolate are anecdotal. This research evaluated the melanging step at three different temperatures in order to determine, via Head Space Solid Phase Micro Extraction (SPME) coupled with GC-MS, which volatile flavor-active compounds are formed or lost during the melanging process and how the time and temperature affect the volatile composition of the chocolate mass. Univariate statistical analysis on the 36 samples (3 temperatures, 4 time points, and 3 technical replicates) was performed using analysis of variance (ANOVA) and multivariate statistical analysis was performed using principal component analysis (PCA) and orthogonal partial least squares (OPLS). These analyses found that time, rather than temperature, accounted for most of the variation among samples. Some trends among chemical classes were identified, and certain indicator compounds were highlighted for their particular relevance to the system. Pyrazines displayed the strongest pattern of any annotated chemical class. As a rule, these compounds started with moderate to high abundance at 0 hours and decreased to extremely low levels by 24 hours. However, most classes of chemical compounds analyzed in this study did not follow strong patterns of behavior predicted by chemical class. Particle size and pH were also measured for each sample. Majority particle size was found to be below 25 microns generally at all time points beyond 8 hours, and pH for all samples had a mean of 5.25 with a standard deviation of 0.177. Analysis showed significant temperature-dependent p-values for several compounds, but significant time-dependent p-values were apparent for a greater number of compounds. For compounds which showed dependency on both time and temperature, the time-dependent p-value was much smaller in most cases. Both PCA and OPLS analyses suggested the same trends. It was determined first, that melanging is a viable method to carry out successful chocolate refining. The final flavor of chocolate made in a melanger is far more dependent on time spent in the melanger than on the temperature of the chocolate system. In addition, classes of chemical compounds did not behave according to unified patterns of behavior in melanging systems (with the exception of nitrogen heterocycles, which decreased in relative abundance). Finally, literature previously published on conching may be successfully extrapolated to melanging.