The Effects of Fermentation and Drying Methods of Theobroma cacao on Quality and Flavor Characteristics

Abstract

Cacao growers in Hawaiʻi face many challenges related to post-harvest processing. This is primarily due to a lack of industry-accepted standards for fermentation and drying, which are critical stages of flavor development within the bean. Farmers in Hawaiʻi have expressed interest in adopting alternative processing methods to improve the quality and marketability of their product. The objective of this study was to evaluate three standard turning intervals for box fermentations, as well as 12 different drying treatments, for their effects on postharvest parameters and quality characteristics of chocolate. Fermentation F211, which was turned initially after 48 hours, followed by 24-hour intervals until completion. Protocols for fermentation and drying were considered best practice for Hawaii. Temperature readings of the top, middle, and bottom layers of each treatment were taken at ten-minute intervals throughout the fermentation cycle. pH values of the cotyledon were measured before and after fermentation, and after drying. Color attributes (L*C*H*) were measured after drying. Sensory evaluation of chocolate, made from dried bean samples of each treatment, were conducted in two parts: evaluation of various flavor intensities (quantitative), and overall preference scores (qualitative). Chocolate samples made from each treatment were compared to a Ghanaian chocolate sample (GS), which acted as a control for flavor. Treatments were conducted monthly over an eight-month period and volume of each fermentation treatment was 227 kg. None of the response variables were shown to have interactions with season. Mean pH of the cotyledon before fermentation was 5.90. pH values decreased among all treatments during fermentation. F222 had the lowest mean among treatments, post fermentation and drying, with values of 4.37, and 5.03, respectively. F222 took significantly longer than F111 to reach the critical fermentation temperature of 43.5 °C, for each layer of the fermentation mass, but there were no differences in mean times spent at or above this temperature between treatments. Mean maximum temperatures were significantly higher for F111 throughout each layer, although there were no differences between treatments in the time it took to reach maximum temperatures. Results from the sensory evaluation by Dandelion Chocolate showed that F222 had the highest score for fresh fruit intensity, and that GS received the highest spice intensity score. There were also differences in mean overall preference scores among treatments. F222, and F211 were scored most favorably, whereas F111, and GS scored least favorably among evaluators. F222 was shown to have consistent levels of mold infestation during fermentation, especially in the bottom corners of the mass. Therefore F211, although receiving a slightly lower preference score, could be recommended to growers as a more reliable alternative to maintain quality. This study also examined the drying behavior of fermented cacao beans subjected to 12 different drying treatments which were categorized by heat source: 1) sun drying; 2) oven drying; and 3) dehumidification drying. Sun drying treatments were conducted at four different sites on Hawaiʻi Island (Pāpaiʻkou, Pepeʻekeo, Kainaliu, and Kawaihae) that represent a gradient of decreasing humidity and rainfall. Mechanical oven drying, and dehumidification drying were both conducted in controlled indoor environments in Hilo. Frequency of the drying interval was either constant or intermittent at each drying location. Moisture content was measured twice daily throughout the drying period. Color attributes and pH were measured before and after drying. Bean samples from each treatment were sent to Dandelion Chocolate Company for an in-depth sensory evaluation of chocolate made from each sample. Response variables were not shown to have an interaction with season. Initial mean moisture content of beans was 54.9± 2.5% wb. There were significant differences in mean drying rates between treatments, with sun drying taking 17.9 days at Pāpaiʻkou (the highest humidity site) and only 2.9 days for a constant oven drying treatment (ODC). Mean starting pH for the testa and cotyledon was 4.54 and 4.7, respectively. A constant sun drying treatment in Pepeʻekeo (SDPC) had the highest pH value for the testa and cotyledon, at 6.2 and 5.6 respectively. Whereas a constant dehumidification drying treatment (DDC) had the lowest testa pH (5.5), and ODC had the lowest cotyledon pH (4.8). Results from sensory evaluations indicated that an artificial intermittent treatment (ODS) had the highest rating for fresh fruit intensity, whereas the control (GS) and DDSD had by far the lowest. Herbal/floral intensity ratings also differed between treatments, with the control having the highest rating, and a natural intermittent treatment (SDKS) having the lowest rating. Mean overall preference scores were not shown to differ between treatments.