Fatty acid profile and solid fat content of Peruvian cacao for optimal production of trade chocolate

Este trabajo fue recibido el 27 de diciembre de 2018. Aceptado con modificaciones: 28 de mayo de 2019. Aceptado para ser publicado: 01 de agosto de 2019. ABSTRACT Using technical procedures, the fatty acid (FA) profile and solid fat content (SFC) of the Peruvian cultivar cacao beans CCN 51 and ICS 6 and the "optimal chocolate", obtained from the mixture of the first two, were determined to assess their quality. These cacao beans were found to have important nutritional values. The FA profile of the cacao beans were similar (p>0.05); however, in the FA profile, the ‘optimal chocolate’ had significant differences (p≤0.05) in terms of palmitic, arachidic and linolenic acid. The n6:n3 ratio for "optimal chocolate" was 12.0 ± 1.7. Cacao beans had the same SFC, and SFC was highly temperature dependent, as determined using a mathematical model for chocolate. The SFC of chocolate refers to hard cacao butter content at temperatures between 20 and 25°C, and solid fat was heat resistant from 25 to 30°C, which is considered valuable in trade chocolate production. The quality-related properties of these lipid fractions imparted nutritional and physical aspects to the optimal dark chocolate for human consumption.


INTRODUCTION
Cacao beans are obtained from cacao pods, the fruit of the cacao tree (Theobroma cacao L.) which grows in all tropical zones of Peru. Cacao butter, constituting 50%-57% of a dry cacao bean, is an important part of the bean owing to its various physical, chemical and sensorial properties .
Identifying the type and quantity of fatty acids (FAs) in cacao butter through research and development laboratories Fatty acid profile and solid fat content of Peruvian cacao for optimal production of trade chocolate Composicion de acidos grasos y contenido de solidos grasos de cacao peruano para la produccion de  Solid fat content (SFC) is a parameter that expresses the solid to liquid mass ratio of fat at different temperatures. The SFC affects physical properties such as consistency, stability and some important sensory attributes such as flavor, aroma and overall acceptability 5 . The SFC from 20°C to 25°C qualifies the hardness of cacao butter. The range of temperatures wherein decrease in the SFC is evident represents resistance to heat, whereas rapid fusion from 32°C to 35°C is responsible for cooling and a creamy sensation during tasting. To evaluate the quality of cacao butter, an important and practical parameter used in the industry is the difference between the SFC at 25°C and 35°C (i.e. ΔS). The presence of solid fat at a temperature below 35°C is recognised as a waxy sensation, which is easily detected during tasting 5 .
Apart from providing sweetness, chocolate offers positive health benefits that should be further investigated 6 . This study aimed to determine the FA profiles, ratios related to healthy nutrition and SFC of two Peruvian cacao cultivars CCN-51 and ICS-6 and the "optimal chocolate".

MATERIALS AND METHODS
This research was conducted at a processing plant in Chocomuseo (Lima, Peru) for cacao beans, Cacao Valley (Lima, Peru) for optimal chocolate production and University of Wisconsin-Madison (Madison, Wisconsin, USA) for the analysis of lipid fraction and determination of physical properties.

Analytical methods
Proximate analysis: The basic chemical composition i.e. moisture, ash, proteins and total fat of unroasted cacao beans and the "optimal chocolate" was analysed in triplicate following the AOAC method for cacao beans and their products 7 . Additionally, cacao paste from the beans was analysed in triplicate for moisture and fat composition 7 .
Caloric content was calculated by proximal analysis 8 expressed in calories per 100 grams of product.
Determination of FAs: The lipid fractions of the cacao and "optimal chocolate" pastes were extracted 7 . Subsequently, using the ISO 12966-1 9 methodology, we prepared methyl esters of FAs 10 . The results are reported as a relative percentage of total FA for the cacao and chocolate pastes.
Determination of SFC: We used the IUPAC Method 2.150b 11 with nuclear magnetic resonance to determine SFC. Lipid fraction of the cacao and "optimal chocolate"pastes were extracted with petroleum ether solvent; the fraction was melted, homogenised and filtered, and 2 mL solutions were measured in test tube at 80°C for 30 min. The samples were then placed in a water bath maintained at 0°C for 90 ± 5 min, passed to a bath maintained at 26°C for 40 ± 0.5 h, placed in a bath maintained at 0°C for 90 ± 5 min and, finally, passed to a bath maintained at the following temperatures for 60 min: 0°C, 5°C, 10°C, 15°C, 20°C, 25°C, 30°C, 35°C and 40°C.

Experimental methodology
At a processing plant, CCN 51 (Colección Castro Naranjal) and ICS 6 (Imperial Collage Selection, District of Uchiza, Peru) were processed (roasted, peeled and ground) to obtain cacao paste. The paste samples with their replicates were evaluated and lipid fractions were extracted to evaluate FA profile. Subsequently, nine different samples of chocolate with two replicates were derived from the mixtures of cacao beans (ICS-6:CCN-51 ratios of 1:9, 5:5, and 9:1) and three dark chocolate formulations (60%, 70% and 80% cacao) were processed under the same traditional processing conditions. In addition, physical properties for chocolate production were evaluated (University of Wisconsin-Madison, USA) to determine the "optimal chocolate" 12 .
Using the surface response method, chocolate was obtained using the mixture with desired physical properties. For validation, the "optimal chocolate" (ICS 6:CCN 51 ratio of 1:9 and 70% cacao) was manufactured again. Fat of the "optimal chocolate" was extracted to evaluate FA profile, calculate the S/U and n6:n3 ratios and to determine SFC (University of Wisconsin-Madison, USA).

Statistical analysis
Experimental values obtained for each evaluated variable (FA, SFC, SAFA, MUFA, PUFA, fat, protein, ash, moisture, crude fibre) were expressed as mean ± standard deviation. The differences of these values between CCN 51, ICS 6 and optimal chocolate were determined by analysis of variance (p≤0.05) and Student-t test for moisture and fat content of cacao paste obtained from CCN 51 and ICS 6. A simple regression method was applied to investigate the relationship between SFC profile and temperature. All statistical analyses were performed using the STATGRAPHICS Plus ® programme.

RESULTS
The proximal chemical composition of cacao beans (Table 1) showed a fat content of 43.6% ± 0.12% and 45.8% ± 0.05% for CCN 51 and ICS 6, respectively. In paste, a fat content of 49.35% ± 0.30% and 52.50% ± 0.11% on a wet basis was observed and 50.43% ± 0.31% and 53.01% ± 0.11% fat content on a dry basis, respectively. Lastly, the proximal analysis of the 'optimal chocolate' paste (CCN 51:ICS 6 ratio of 1:9, 70% cacao) defined on a wet basis showed a moisture and fat content of 1.60% ± 0.22% and 34.56% ± 0.03% (Table  1) and a fat content on a dry basis of 35.11% ± 0.03%; this helped in determining the integral lipid fraction of the FA profile.
The FA profile and SFC of the cacao butter from the cacao beans and "optimal chocolate" are shown in table 2 and table  3. Values were expressed at 100% of the lipid composition to compare the values of FA among the countries in South America and globally and shows the ratios of S/U and n6:n3 for the "optimal chocolate". The SFC models of cacao butter from the cacao beans and 'optimal chocolate' are shown in figure 1.   Notably, the SAFA content (63.80% ± 0.27% and 63.65% ± 0.69%, respectively) was higher than the UFA content (35.93% ± 0.51% and 36.19% ± 0.81%, respectively). Regarding stearic FA (C18:0), a neutral saturated FA type 2 , we found the values of 32.77% ± 0.35% and 33.87% ± 0.31%, respectively, similar to those reported for Colombian varieties (31.9%-36.6% 13 ).
The S/U ratios (ratio of saturated FA to unsaturated FA) for CCN 51 and ICS 6 were similar between the two varieties (p > 0.05), with the values of 1.78 ± 0.03 and 1.76 ± 0.06, respectively, whereas the ratios for the varieties found in Ghana and Ecuador is 1.72 ± 0.05 and 1.65 ± 0.04, respectively 2 indicating that the SAFA content is higher than the UFA content. Notably, the S/U ratio for the beans found in Ecuador is the lowest of all. Perea et al. 13 report that a high S/U ratio classifies butter as hard, which is desired by the food industry. In this study, the CCN 51 and ICS 6 varieties obtained remarkably higher S/U values.
The lipid fraction of the "optimal chocolate" was 34.56% (Table 1). In previous reports, Mursu et al. 16 used dark chocolate with 33.00%, Salinas and Bolivar 17 used bar-type chocolate (drinking) with 37.00% fat and Chire et al. 18 used dark chocolate with 36.11% fat. Because the 'optimal chocolate' had a high fat and energy content (34.56%), daily consumption should be regulated to 25 g/ d 16,2 . In terms of fat content (35.11% ± 0.03%, dry basis), the FA profile of the "optimal chocolate" was determined. While the SAFA content of the "optimal chocolate" was 61.90% ± 1.01% that of the chocolate prepared with the cacao beans from Ghana and Ecuador is 66.03% ± 2.09% and 63.95% ± 1.39%, respectively 2 , and that of the bartype chocolate (drinking) reaches 63.5% 17 . For the "optimal chocolate", the stearic FA content (C18:0) was 33.34% ± 1.93% representing one third of the fat content. Stearic FA increases high density lipoprotein cholesterol and serum triglyceride levels in blood 2 . PUFAs were present in high contents in the "optimal chocolate" (4.36% ± 1.01%) compared with those in the dark chocolates from Ghana and Ecuador (1.78% ± 0.48% and 2.36% ± 0.59%, respectively 2 ; 2.45 and 1.85 fold more, respectively).
As per previous studies, dark chocolate traded in Finland has 3.03% linoleic and linolenic FAs 16 , bar-type chocolate (drinking) from Venezuela has 3.16% 17 and the "optimal chocolate" from Peru (this study) had 4.35%. The S/U ratio is a health indicator for chocolate with standard fat content: the S/U ratio for 'optimal chocolate' reached 1.67% ± 0.03%, almost similar to that of bar-type chocolate (1.75% 17 ). In the study by Torres-Moreno et al. 2 , the S/U ratios for chocolates from Ecuador and Ghana are 1.77 ± 0.01 and 1.94 ± 0.06, respectively, which increase during cocoa bean processing and chocolate production, with inverse effects on the results of the present study, because the S/U ratios for cacao beans (1.78 ± 0.03 and 1.76 ± 0.06) decreased to obtain the "optimal chocolate" (1.67 ± 0.03).
In addition, essential FAs (omega 6 and omega 3) produce metabolites that exert positive and negative effects on health and if the omega 6:omega 3 ratio is higher, the product is injurious to health 19 . The "optimal chocolate" had 4.35% of omega 6 and omega 3 2 , whereas as chocolate from Ecuador has 2.30% and chocolate from Ghana has 1.80%. Per Mursu et. al. 16 , chocolate traded in Finland has 1.00% of omega 6 and omega 3. Thus, our results indicate a good presence of essential FAs in the chocolate. Because the ratio of omega 6:omega 3 ( Table 2) for "optimal chocolate" was 12.0 ± 1.7, a daily diet must be supplemented with food products that help attain a ratio of essential FA of 5:1 19 . Chocolate from Ecuador has 21.6, 10.5 2 from Ghana and 9.0 16 from Finland. Thus, our results of 12.0 was near the ratio from Ghana. This could be due to the origin or processing factors. Curti et al. 20 studied the debittering process of Lupinus species and show an omega 6:omega 3 ratio from 10.3 ± 0.3 to 8.8 ± 0.6. According to Mataix 21 , the FA content in 100 g of food comprising various fruits and nuts, i.e. almonds, peanuts, walnuts and "sacha inchi" seeds 22 , has omega 6:omega 3 ratios of 37.78, 36.43, 5.26 and 0.48, respectively, indicating that only walnuts possess a balanced value.
It is recommended to characterize cocoa butter of other varieties and countries and differentiate them by zones. Innovative processes should be developed to improve the nutritional indicators of cocoa lipids.
The SFC of the lipid fractions of the CCN 51 cacao, ICS 6 cacao and the "optimal chocolate" ( Table 3) from 0°C to 40°C with intervals of 5°C showed that the curves decreased and were characteristic of cacao. CCN 51, ICS 6 and the "optimal chocolate", presented no significant differences in terms of SFC. Using the same method as used in the present study (IUPAC method 2.150b), Wennermark et al. 23  Similarly, for a type of cacao butter from Ghana, Tran et al. 25 report a SFC of 85.90% ± 0.20% at 10°C, 74.60% ± 0.20% at 20°C, 65.30% ± 0.30% at 25°C, 40.30% ± 0.50% at 30°C and 0.70% ± 0.20% at 35°C, which are similar to the results of the present study at temperatures of <20°C. However, at temperatures of 25°C, the SFC was higher in our study, confirming the heat resistance of cacao butter. In the ΔS experiment, the SFC values in CCN 51, ICS 6 and the "optimal chocolate" were 65.01%, 65.52% and 65.34%, respectively, which were within the study range reported by Tran et al. 25 In the last section of the SFC curve (temperatures of 35°C-40°C), the butter in the present study had waxy residues, more than that of the varieties from Vietnam and Ghana 25 .
Remarkably, CCN 51, ICS 6 and the "optimal chocolate" presented an inverse relationship between their SFC and temperature (R2 = 96.61%, 96.81% and 96.55% respectively), which are empirical models based on observation and experimentation 26 . Data were analysed by simple regression, where the dependent variable was SFC and the independent variable was temperature and the model was double square. The mathematical models for the butter (lipid fraction) obtained for CCN 51, ICS 6 and the "optimal chocolate" (Figure 1) had negative slopes and were quadratic.
The results showed that the SFC presents a hard cacao butter, resistant to heat and with serosity residues. Therefore, to characterize the SFC of the cacao butter a mathematical model that can confirm the quality of the cacao butter is recommended. For example, for the products with cocoa butter replacers or substitutes, the SFC should be different.

CONCLUSIONS
The FA profiles of the Peruvian cacao varieties CCN 51 and ICS 6 from the district of Uchiza (Peru) showed nutritional potential in terms of PUFA content (3.39% ± 0,48% and 3.29% ± 0.39%, respectively) that was higher than that of the African varieties. Thus, the Peruvian varieties provide a greater nutritional contribution to the "optimal chocolate" blend processed with 70% cacao and cacao bean mixes (10% CCN 51 and 90% ICS 6). The PUFA content of the "optimal chocolate" was 4.36% ± 1.01%, a level significantly higher than that of other chocolate producers in the world and those recognized in Ecuador. Similary, the 'S/U ratios' calculated for CCN 51 cacao, ICS 6 cacao and the "optimal chocolate" were 1.78% ± 0.03%, 1.76% ± 0.06% and 1.67% ± 0.03%, respectively. The n6:n3 ratio for the "optimal chocolate" was 12.0 ± 1.7, thus differentiating it from those obtained from African cacao.
The high SFC at low 10°C and medium 20°C temperatures is characteristic of hard cacao butter, and a significantly high SFC at high temperatures (35°C and 40°C) is considered beneficial for the distribution of products in tropical areas because it imparts heat resistance. The SFC in terms of temperature had a negative slope formed a quadratic equation.