Starter Yeasts Impact Flavour Precursors in Brazilian Cacao
The following is a summary of the stated research mentioned above. The content summarized here, including the figures and tables, all belong to the researchers (unless otherwise indicated). The summary attempts to stay as close to to the original paper as much as possible with some adjustments in regards to jargon, length, or to focus on bean to bar aspects
Introduction
The purpose of this work was to see how using starter yeasts to ferment cacao would influence the properties and flavour of the cacao. Normally cacao is allowed to ferment spontaneously, from yeasts and bacteria from the surrounding environment. However, some inoculate fermentation heaps with specific strains in order to have more control the direction the fermentation takes, and it’s impact on cacao quality.
Fermenting the cacao seed & fruit together is a crucial step in developing good cocoa flavour and precursor aromas later developed during roasting.
The results showed that various species of yeasts had various impacts
Materials & Methods
Yeast Isolation & Screening
The yeast was collected from cocoa fermentations during the early crop season in Leolinda farm (Urucuca -BA, Brazil). The mucilaginous part was used to isolate microorganisms and plating them on Sabouraud agar containing chloramphenicol. The plates were incubated form 72 hours at 25*C, and the yeast colonies were then isolated. Of all the isolates, 112 were randomly chosen for enzymatic tests.
Fermentation
The experiment took place in either September (using 4 of the 8 strains of yeasts), or December (using the other 4 of the 8 strains of yeasts). Each batch of 4 also had a 5th batch used as the control.
The seeds were homogenized and then separated into 15 lots, and placed in wooden boxes of 45 kg each. The boxes were then inoculated with 3.26x108 yeast cells per kilogram of beans. The fermentation proceeded for 144 hours, moving the mass every 48 hours. At every 12 hours, the temperature of the mass was taken, and 200g of almonds were removed for physiochemical analysis. The fermentations were performed in triplicate for each yeast strain tested. A box containing no inoculum (natural fermentation) was used as a control.
After fermentation, the cacao was dried on wooden platforms (barge system) until reaching 6-8% humidity.
Throughout the process of curing and preparing the beans, they were tested for the following:
pH of the pulp, and pH of the dry beans
Titratable acidity of the beans
Cut test
Analysis of reducing sugars (higher levels of reducing sugars may indicate higher chances of more aromas forming)
Analysis of free amino acids (higher levels of free amino acids may indicate higher chances of more aromas forming)
All experiments were performed in triplicate, and a variance analysis (ANOVA) and a Turkey’s multiple comparison test were applied to determine significance of the values between those masses inoculated and the control.
Results
Selection of yeasts to use as inoculum starter
Eight yeast strains were selected from the original fermentation from Leolinda farm (Table 1). They were chosen to see if the inoculum influenced obtaining higher values of flavor precursors.
Fermentation temperature
There was no significant difference between the control and experiment fermentation mass temperatures. In the September batch (yeasts A-D) the average temperatures started at 24*C, peaked at 45.2*C, leveled at 43*C, and declined to 37.2*C. In the December batch, (yeasts E-H) the average temperatures started at 27.2*C, peaked at 45.*C and remained there until it declined to 34.6*C
The pH
September batch increased from 3.59 to 5.6, while the December batch increased from 4.1 to 4.9.
pH, acidity, and cut test of beans
Inoculums A (R. mucilaginosa), B (T. delbrueckii) a significantly lower pH than the control 1. Inoculum G (Saccharomyces cerevisiae) had a significant higher pH than control 2.
The titratable acidity of the experimental seeds were statistically lower than the controls (Figure 1). The seeds inoculated with Candida parapsilosis (C) had the lowest acidity.
Titratable acidity.
The cutting test showed uniformity of well fermented seeds, and brown colour of more than 95% in all batches.
Cut test results. C1 is the control for experiments A-D, and C2 is the control for experiments E-H.
Analysis of Reducing Sugars
The level of reducing sugars were significantly higher in batches inoculated with T. delbrueckii (B) and C. parapsilosis (C), both of which were higher than their control.
Experimental batches F, G, and H (I. orientalis, S. cerevisiae, and P. membranifaciens) showed lower levels of reducing sugars compared to their control.
Analysis of Free Peptides and Amino Acids
Batches A-D were all had levels of free amino acids that were higher than the control in the first fermentation period. In the second fermentation period, only E had higher levels of free amino acids than the control. This can be see in Table 3.
Discussion
Temperature and pH didn’t seem to be significantly different between experimental inoculated batches and the control.
The most important acids, such as lactic and acetic acid, are produced by the degradation of sugar by the microbes within the fruit pulp. Malic, tartaric, oxalic, citric, and phosphoric acid are less important to the final pH of the cacao beans. The final pH is influenced much more by the sugar content of the pulp, as opposed to whether the batch was inoculated or not.
This acidic environment, however, is important in that it activates enzymes in the cotyledon of the cacao bean, resulting in higher levels of reducing sugars and amino acids.
Titratable Acidity (Fig 1)
Another way to measure the acidity, is through detecting titratable acidity (Fig. 1). Although you want to increase acidity during fermentation in order to develop flavour pre-cursor molecules, you don’t want your cacao beans to be too acidic in taste. All the inoculated experiments A to H had a lower titratable acidity than their controls. Here the researchers connect this to “increasing the quality” of the beans.
Protease and Pectinase (Table 1)
Pectinase breaks down pectin, a polysaccharide, into reducing sugars. Protease breaks down proteins into free amino acids. Both reducing sugars and free amino acids, which are formed during fermentation, combine to form aroma molecules during roasting. Therefore, higher levels of these may lead to higher levels of aroma molecules in cacao.
Inoculum C (C. parapsilosis) resulted in the lowest acidity overall. The researchers suggest this is due to its pectinolytic activity (which aerates the mass more and promotes a reduction in lactic acid bacteria). You can see in Table 1 above, that C. parapsilosis had the greatest level of pectinase.
Inoculum B (T. delbrueckii) generated the highest concentration of free amino acids (Table 1), likely due to it’s proteolytic activity.
The researchers suggest that these inoculants can contribute to the change in concentration of flavour precursors (reducing sugars and free amino acids).
