Introduction

Cocoa and dark chocolate products contain much more polyphenol content compared to other products such as wine, tea, and fruits which are often touted for their high levels of antioxidants. Image by @helenayankovska

Cocoa has been used for millennia, and documented as far back as approximately 5500 years ago from a site in Southern Ecuador called Santa Ana-La Florida (1,2). Most our knowledge of pre-Columbian chocolate was from the Mayans and Aztecs who consumed it as a drink (3).

There have been many curative and benefits of consuming chocolate and cocoa products for generations, but only recently do we have some of these positive effects scientifically validated. Improving analytical technology and protocols have allowed for this. Many of the clinical studies in recent years have pointed to the presence of antioxidant compounds in cacao/chocolate as the source of these benefits (4).

The most represented antioxidant compounds present in cocoa are catechins, anthocyanins, and proanthocyanins, with (-)-epicatechin being the most abundant flavanol. They have been associated with reducing blood pressure and glycemia during pregnancy (6), improving cognitive function, reducing blood pressure in the elderly with mild cognitive impairment (7), disease modifying properties in patients affected by Alzheimer’s disease (8), health effects against oxidative stress, chronic inflammation, risk factors for cancer, and other chronic diseases (9).

In 2012, the European Food Safety Authority issued a health claim stating the consumption of 200 mg/day of cocoa flavanols (found in 2.5 g of high-flavanol cocoa powder or 10 g of high-flavanol dark chocolate), consumed in the context of a balanced diet, helps to maintain endothelium-dependent vasodilatation (contributing to normal blood flow) (10).

Cocoa contains very high levels of polyphenols, more so than other crops, foods, and beverages. Even tea and red wine have lower levels of polyphenols (11, 12). However the levels of polyphenols in cocoa can fluctuate a great deal depending on factors such as their genetics, growing conditions, pod storage, fermentation (13), drying (14), alkalization, roasting, and conching (the last three being the most detrimental to cocoa polyphenol levels (15-20)). Alkalization is used mostly in cocoa powder production, and conching can be avoided, but roasting is normally always performed. Both roasting and fermentation are stops which contribute a great deal to the overall bouquet of aromas in cacao and chocolate, but are also detrimental to cocoa polyphenols and account for over 50% loss of their loss (21).

Since high levels of polyphenols in cocoa products are favorable, an understanding of the process of cocoa/chocolate manufacturing could benefit in preserving these compounds. Therefore, here cocoa beans were roasted using three different roast temperatures in order to produce Modica chocolate (Cioccolato di Modica). Modica chocolate is considered to have a mild production process (cold mixing of the cocoa paste and sugar) and does not mix at high temperature/conching (22). The samples were analyzed at each step for their polyphenol content, epicatechin concentration, antioxidant capacity, and mean degree of polymerization of epicatechin oligomers.

Methods

Cocoa beans were refined into cocoa liquor (cocoa paste) using equipment produced by the SELMI Group (Santa Vittoria d’Alba, Italy). Cocoa beans were roasted in a Roaster 106 at three different temperatures (Table 1), and repeated three times for reach roasting temperature. The beans were then winnowed and ground to a particle size between 200 and 250 um (Stainless steel roller crusher, Grinder Mill, SELMI Group, Santa Vittoria d’Alba, Italy), and then refined to a particle size of 20-25 um via Micron ball refiner by SELMI group. This was then combined with sugar to obtain Modica chocolate. This type of chocolate was chosen due to the mild processing technique and less thermal stress. The Modica chocolate was prepared by blending molten cocoa liquor at about 50 degrees Celsius with sugar in a 60/40 ratio.

Total polyphenol content (TPC) was determined through the Folin-Ciocalteu assay and all samples were analyzed in triplicate. There are two main mechanisms at the basis of antioxidant action of phenolic compounds, either through an electron transfer or a hydrogen transfer process (or combination of both) (23,24), therefore in this study the comparison among the total antioxidant capacity (TAC) of each extract was determined through ferric reducing antioxidant power (FRAP) assay. To quantify the epicatechin, a high-performance liquid chromatography analysis (HPLC-UV) was carried out as well as a gas chromatography-mass spectrometry analysis (GC-MS). Pearson’s correlation test was performed to determine the correlation between epicatechin content and the TPC or TAC.

Results & Discussion

TPC and TAC

The samples, roasted at 3 different temperatures, were analyzed for their TPC, TAC, epicatechin content, and mean degree of polymerization (mDP). The 7 stages of samples were: unroasted cocoa beans, roasted cocoa beans, cocoa nibs, cocoa liquor, refined cocoa liquor, and final chocolate. It’s interesting to note that measured polyphenol content (TPC) of the roasted cocoa beans, compared to unroasted beans, actually increased a great deal in the 100 and 130 degree Celsius programs (Figure 4), but then a reduction in TPC in the successive processing steps. The TPC of beans roasted at 80 degrees Celsius remained relatively the same even after roasting. Keep in mind also that the chocolate polyphenol content here referred to the grams of refined cocoa liquor, and so there is no dilution effect due to the added sugar (32). No other ingredients were added.

A similar trend was also observed for TAC (Total antioxidant capacity), with the highest increase after roasting at 100 and 130 degrees Celsius, very similar to TPC data. In the 80 degree roasting program, the final product had a slightly higher TAC than the cocoa beans before roasting (Figure 4).

Suazo et al. (34) found the higher the roasting temperature or longer the duration, the greater the loss of polyphenols. However, as shown here, the technique and protocol employed can be optimized to reduce the detrimental effects on polyphenols. For example, superheated steam roasting has shown to lower the impact on polyphenols than conventional roasting (35). The air velocity (speed) and humidity during roasting have been observed to impact the polyphenols (36).

Keep in mind that a reduction in TPC is not always followed by a reduction in TAC, which can even increase as TPC decreases (32). Although the total polyphenol content may decrease during roasting, fractions of reducing substances may increase. And depending on the method used to quantify the free-radical scavenging activity (to quantify the TAC), these same fractions may contribute to the TAC (37).

Comparing data from various literature is not as straight forward as many have different experimental setups. That said, some authors have reported increase in TPC when roasting is performed at low temperatures (38,39). The experiment here differed from others, as others used temperature “ramps”, while here the 3 temperature points were fixed for the duration of the roasting. In this experiment, the intermediate temperature programs (100 degrees Celsius) saw the highest increase in polyphenols/antioxidants, but overall, the best preservation for antioxidant capacity in the final product was seen at the lowest temperature (80 degrees Celsius).

This rise in phenols/antioxidants may be due to new-born compounds, other than polyphenols, to the total polyphenol content due to the non-complete specificity of the polyphenol quantification assay. The Total phenolic content was obtained by using the Folin-Ciocalteu assay, which is a quick and cheap method to quantify polyphenols in foods. The reaction of this FC assay produces blue complexes which are then quantified via an spectrophotometer at a wavelength around 760 nm (40). The problem is that the Folin-Ciocalteau regent may react with other compounds, giving rise to an underestimation or overestimation of the phenolic fraction. It has even been suggested that the Folin-Ciocalteau should be used to measure TAC rather than total phenolic content (42).

MRPs

There is also the possibility that the Maillard reaction products (MRPs) formed during roasting also contribute to the compounds measured. During roasting or other thermal treatment, non-enzymatic reactions between reducing sugars and amino acids, peptides or proteins takes place and new compounds (referred to as MRPs) are created. Melanoidins, high-molecular weight brown polymers, are a class of MRPs which contribute to the colour and flavour development of cocoa. It is known that Melanoidins react with the Folin-Ciocalteau reagent and possess antioxidant activity. Therefore, the changes in both TPC and TAC values are due to these melanoidins and possibly other MRPs developed during roasting or even grinding/refining (which also creates temperature increase) (43-35). As well, cocoa melanoidins may contain polyphenols such as epicatechins either covalently bound in their backbone or adsorbed in them. When these compounds had undergone partial hydrolysis, the melanoidin-bound phenolic compounds maintained their scavenging properties (46).

Epicatechin

Since epicatechin is the most abundant antioxidant found in cocoa, this was the one chosen to be analyzed throughout the experiment (Figure 5). Epicatechin increased during the production process when roasting at 80 degrees Celsius, following a trend similar to TAC. A clear trend for epicatechin levels during roasting at 100 and 130 degrees Celsius could not be observed.

TPC, TAC, and epicatechin correlations

A Pearson correlation between epicatechin and TPC, as well as epicatechin and TAC was observed for the 80 and 100 degree roasts. No correlation was observed for the 130 degree Celsius roasting program between epicatechin and TPC/TAC (Figure 6), which confirms that MRPs (Maillard reaction products) played a role in TPC and TAC for the 130 degree samples. This suggests that at low and intermediate roasting temperatures, epicatechin is what contributes to the antioxidant activity of cocoa polyphenols, but that at high roasting temperatures other compounds other than epicatechin contribute to the antioxidant capacity.

The trend overall including roasting and other processes is that the polyphenol content (TPC) and antioxidant capacity (TAC) increased and then decreased to various levels depending on the stage and the roast. However, in the samples from the lowest roasting temperature (80 degrees Celsius) the epicatechin levels remained relatively the consistent. While TPC and TAC levels are indirect measurements, but epicatechin concentration has been quantified directly by HPLC (High-performance liquid chromatography). Therefore, the increase in epicatechin concentration during cocoa production is quite reliable, due to the release of epicatechin molecules from epicatechin oligomers as a consequence of roasting or all the other steps involved.

Therefore, oligomers were extracted and analyzed to calculate their mDp (Figure 5). Mean degree of polymerization (mDP) is essentially a measure of the number of repeat units in a polymer or oligomer (oligomer are smaller than polymers, usually only 10 chains or less). A decrease in mDP value during the different steps of chocolate production would indicate a reduction of the polymerization degree of the oligomers (oligomers breaking apart into more yet smaller chains). This may indicate the release and/or rearrangement of monomers, such as epicatechin. It is speculated this is what is responsible for the epicatechin variation. For example, cocoa samples under the mildest temperature program (80 degrees C) hand an increase in epicatechin corresponding to a decrease in mDP (Figures 5 and 7). The same behaviour was not observed in the two higher roasting programs.

Catechins

Catechins oligomers, known as proanthocyanidins, have their own antioxidant capacity (47). Their content was not evaluated here, so only the mDP is reported, therefore it is not possible to draw conclusions on their contribution the TAC of the samples. As stated, it’s not common to see a rise in polyphenol content during roasting or processing, but some reports observe an increase due to hydrolysis of oligomers (48, 49). De Taeye et al. (50) found evidence of the formation of new oligomers during roasting and other steps of chocolate processing, which wasn’t present from the start. They found dimmers and trimers (chains of 2 or 3 monomers) in the post roasted/processed products, but not found in the unroasted cocoa beans. This suggests the possibility of oligomer rearrangement, which may describe the variations in mDP at different steps of the process.

PCA plots

A PCA plot was used to show the similarities between the various samples (Figure 8), and groups similar ones together . Here, PC1 relating to the polyphenols content and antioxidant activity, and PC2 related to oligomerization degree. As shown in Figure 8, the lowest roasting temperature (80 degrees Celsius) are characterized by high values in PC1 and PC2 (top right corner in blue), which indicate an increase in polyphenol content and activity. At intermediate temperature (yellow) polyphenol content and activity was high, but not as high as the lowest roasting temperature program. Finally, the highest temperature program (red) are plotted mostly in the lower left, showing a reduction in polyphenol content and activity.

It’s interesting to point out that all three chocolate samples (seen in the top left quadrant) were not grouped together with other samples from their corresponding roasting program.

Conclusion

Roasting and other steps of the chocolate manufacturing process are responsible for variation in levels of polyphenol content. The final product is generally lower than that of the starting raw material (unroasted cocoa beans). However, mild roasting has been shown to preserve many of these healthy compounds, and that real roasting programs can steer the levels of these compounds. Most of the available reports on such matters have focused on a single manufacturing step, or reflect conditions in a lab which may be quite different than those in a real manufacturing facility. Therefore, much more research is required to study the fate of antioxidant compounds during the whole process of chocolate manufacturing. A better understanding of the evolution of polyphenols throughout the process of chocolate making will allow for optimization of the process which in turn will allow to obtain consistent results that may be reproduced during manufacturing.

Effect of Cocoa Roasting on Chocolate Polyphenols Evolution