Effect of Agroforestry and Cocoa Producing Geographical Origin on the Flavor Profile of Beans and End-Chocolates in the Climate Change Context in Côte d’Ivoire

1. Introduction

Cocoa beans and chocolate are known as luxury foods that provide an astringent taste and typical aroma [1]. Cocoa, a perennial crop highly cultivated only in the equatorial regions, holds significant economic importance in several countries including Côte d'Ivoire and Ghana [2], accounting for 70% of the cocoa international supply [3] providing incomes for 2 millions farmers. Unfortunately, both countries are incriminated in deforestation while benefiting from cocoa production [4]. Cocoa bean is the main raw material for chocolate and other cocoa products [5]. Chocolate is one of the most consumed foods worldwide due to its unique flavor and sensory characteristics, resultiig comes from the unique and fascinating cocoa flavor [6,7]. The quality parameters of the finished chocolate are strongly influenced by cocoa farmers' farming practices at the start of the chocolate supply chain [8]. The chocolate's flavor depends on little controllable variables such as the genotype and the agroecological niche, and on the other side, on primary postharvest mainly fermentation and roasting [6]. Different studies and surveys show differences in the farming practices regarding growing between farmers within the same country [8]. However, several studies emphasize that the primary post-harvest processing of cocoa, such as cocoa pod opening delay, fermentation, and drying carried out mainly by farmers, must target the production of specialty cocoa and controlling the processing conditions by integrating the quality characteristics required by the chocolate market [6,9]. Furthermore, the chocolate's flavor depends on other various factors including genotype and age of cocoa tree, soil quality, microclimatic conditions, as well as through the industrial process of beans until obtaining chocolate where the roasting is emphasized ([10,11]. The cocoa postharvest processing is mediated by a dynamic of biochemical reactions under the actions of successive microorganisms including yeasts, lactic acid bacteria, and acetic acid bacteria spontaneously inoculating cocoa pulp, and producing ethanol and lactic and acetic acids ([6,12]. Endogenous enzymes catalyze the production of peptides and amino acids from seed storage proteins [13], and the inversion of sucrose and the subsequent formation of reducing sugars occur [14]. Resulted metabolites from enzymatic reactions of proteolysis and hydrolysis of the biochemical seed components are considered as flavor precursors [6,15]. Then, during roasting, these products interact through nonenzymatic browning Maillard reactions leading to the generating of molecules such as pyrazines, alcohols, ketones, aldehydes, esters that are in turn responsible for final sensory notes comprising the chocolate flavor, e.g. flowery, fruity, caramel, nutty, among others [9]. Thus, the aromatic quality of cocoa beans and the sensory perception of chocolate depend on the biochemical composition of fresh cocoa pulp ([16,17,18]. Cocoa is one of many drivers of deforestation, and assessments must consider competing sectors in a landscape [19]. Several inconveniences of deforestation including reduction of agrarian forest areas, impoverishment of soils, disappearance of biodiversity, and food insecurity for the populations are reported [20]. However, cocoa cultivation that maintains higher proportions of shade trees in a diverse structure (cocoa agroforestry) is progressively being considered as a sustainable land-use practice that meets ecological, biological, and economic objectives [21]. Since a long time, cocoa agroforests have also been regarded as environmentally preferable to other forms of agricultural activities in tropical forest regions [22,23]. By creating favorable microclimatic conditions, Agroforestry is viewed as a strategy to sustainably enhance agricultural production, which includes cocoa [24]. It was previously reported that the shade environment produced in agroforestry practices affects the morphology, anatomy, and chemical composition of intercropped forages and therefore may affect forage quality [25]. Our assumption is that the biochemical composition of both pulp and cocoa beans can be affected by the tree shade favored by agroforestry in cocoa cultivation. It is essential to evaluate the balance between the positive and negative effects of agroforestry on the flavor of cocoa and chocolate, particularly as climate change changes weather patterns. Although Côte d'Ivoire is the world leader producer of raw cocoa beans, up to now, no multiphasic study really has addressed the issue of cocoa bean flavor and chocolate organoleptic quality in relation to agroforestry as a relevant agricultural cropping system. Moreover, raw cocoa beans sourced from this country are not known for their fine aromatic quality. The present work investigated the highlighting the effects of agroforestry as a cocoa cropping system on both the aromatic quality of cocoa beans and the sensory perception of chocolate produced thereof. In response to this this goal, three major research questions were formulated:

i) What could be the effect of agroforestry as a cropping system on the flavor profile of cocoa beans?

ii) Does agroforestry impact the occurrence of desirable flavor compounds in chocolate?

iii) Is the chocolate made from agroforestry cocoa beans more delicious than the ones made from full-sun cocoa systems?

2. Materials and Méthods

2.1. Sites of Carrying Out of Research Works

The harvest of ripe cocoa pods and primary cocoa post-harvest processing (pod opening, fermentation, and sun-drying) was carried out in 5 main producing areas such as Agnibilekrou (East, geographic coordinates : 7° 7' 49.012" N 3° 12' 11.074" W), Adzopé (South East, geographic coordinates : 6° 6' 25.726" N 3° 51' 19.264" W); Divo (South center, geographic coordinates : 5° 49' 59.999" N 5° 22' 0.001" W), Guibéroua (West center, geographic coordinates : 6° 14' 9.805" N 6° 10' 16.536" W), Méagui (South west, geographic coordinates : 5° 24′ 00″ N 6° 34′ 00″ O) on both agroforestry and full-sun system (control) cocoas (Figure 1).

2.2. Materials

Fresh cocoa seeds originating from the mature and full ripe pods of the Ivorian hybrid cultivar, commonly known as "Mercedes" (Amelonado × West African Trinitario), harvested [27] in November 2023 and 2024 from both agroforestry and full sun systems peasant plantation were used in the present work.

3. Methods

3.1. Cocoa Beans Fermentation

The cocoa pods were stored for 3 days before manually opening with a wooden billet as a bludgeon according to the method of Guehi et al. [26]. The fresh cocoa beans were extracted manually without placenta and then sorted to discard the rotten beans. Cocoa beans fermentation process was carried out in wooden boxes in duplicate on the same time at all 5 fermentation sites for 5 days regardless of both cropping system and producing areas. The fermenting cocoa beans were turned simultaneously at 2 and 4 days in all fermentation locations [27] .

3.2. Cocoa Beans Sampling

Three samples of 1000 g of 5-Day-fermented cocoa beans were withdrawn from both batches at each fermentation location. All operating workers wore new sterile gloves for the removal of the fermenting cocoa beans from the heap as previously reported by Koné et al. [28]. To ensure moisture removal of 7-8%, every fermented cocoa bean sample was sun-dried on the rack before carrying out chemical analyses [28]

3.3. Cocoa Volatile Compounds Analysis

Fifty grams of dry fermented cocoa beans were manually deshelled and ground into fine powder (0.5 µm) using a Moulinex grinder (John Gordon®, London, United Kingdom) and then stored at -20 °C in a glass vial hermetically closed [27]. The volatile compounds were extracted from 2.5 g of the cocoa powder from each cocoa bean sample using the headspace solid-phase microextraction technique (HS-SPME) and fibers of 50/30 µm divinylbenzene/carboxene/polydimethylsiloxane (DVB/CAR/PDMS, Supelco, Sigma-Aldrich N.V., Bornem, Bel gium) according to the previously described method by Nascimento et al. [29] but using n-butanol as internal standard. Each volatile compound was identified using three criteria: (i) by comparison of the retention index with the CIRAD aromatic database [30], (ii) by matching their mass spectra with those obtained from a commer cial database (Wiley275.L, HP product no. G1035 A) and (iii) whenever possible, the identification was confirmed using pure standards of the components [31]. The formula used to calculate the concentration of each volatile compound was as follows. :

${\mathrm{q}}_{\mathrm{i}}\left(\mathrm{\mu }\mathrm{g}.{\mathrm{g}}^{-1}\right)={\displaystyle \frac{60\times {\mathrm{A}}_{\mathrm{i}}}{{\mathrm{A}}_{\mathrm{b}\mathrm{u}\mathrm{t}}\times {\mathrm{m}}_i}}$where q_i is the relative concentration of volatile compound; A_i is the area of aroma compound I; A_but is the area of 1-butanol (internal standard); 60 is the concentration of 1-butanol in 100µL of a test sample expressed in µg.L⁻¹; m_e is the mass of the cocoa powder or the chocolate sample introduced into the vial in g.

3.4. Sensory Analysis of Chocolate

Two samples of 2×500g of dry 5-day fermented cocoa beans were taken from each cocoa bean batch treated at five fermentation locations [27]. Whole cocoa beans were roasted at 125°C for a duration of 25 minutes [30]. For the chocolate sensory perception, twelve professional judges, including six women and six men frequently trained, were asked to smell and taste each agroforestry chocolate sample against the full sun chocolate of each cocoa-producing area. They found significant differences in organoleptic attributes of chocolate produced from fermented cocoa bean processed at each tested location of fermentation in comparison to the control. Sixteen sensory descriptors were evaluated simultaneously using a scale ranged from 0 to 10, and a total score for each chocolate sample global quality was assigned.

3.5. Statistical Analysis

The area of the chromatographic peak of each flavor compound precursor was calculated using the software (Instrument Data Analysis) and then exported to Excel. The statistical analyses were carried out with the XLSTAT PLS2 (Addinsoft, New York, USA). Microsoft Excel Program 2013 (Microsoft Corporation, Redmond, Washington, USA) was used to analyze the sensory perception data. Principal Component Analysis (PCA) was used as an unsupervised method to reveal clustering within targeted favor profiles, while Partial Least Squares Discriminant Analysis (PLS-DA) was employed as a supervised method to identify discriminating aroma compounds across cropping systems and geographical origins of cocoa production [31]. The testing of the equality of variances was performed with the Fischer test with a single factor (p < 0.05) in order to indicate the significant differences between volatile compound content of cocoa beans and chocolate samples, the sensory perception of chocolate produced from tested cocoa as affected by the cropping system or the producing geographical origin [27].

3.6. Effects of Agroforestry on the Desirable Flavor Compounds of Dry Fermented Cocoa Beans Different Producing Geopgraphical Origins

Table 3 shows the concentration of several positive flavor compounds of dry fermented cocoa beans regarding both the agroforestry system and the cocoa-producing regions. A total of 15 useful flavor compounds were found in all dry fermented cocoa beans analyzed regardless of the corpping system. The detected desirable flavors belong to various classes such as esters (3-methylbutyl acetate, 2-methylbutyl acetate, Benzyl acetate), aldehydes (2-methylpropanal, 2-methylbutanal, 3-methylbutanal), alcohols (Benzyl alcohol, 2-phenylethanol, linalool), ketones (acetoin, 2-acetoxybutan-3-one, acetophenone), and pyrazines (2,3,5-trimethylpyrazine, tetramethylpyrazine). The fermentation process significantly influences cocoa's flavor and aroma [54]. Several desirable aroma compounds recorded highest concentrations in agroforestry dry fermented cocoa beans from the Guiberoua region. Thus, 2 -methylbutyl acetate, 3-methyl butyl acetate and benzyl acetate were found at higher concentrations of 84.2±3.7, 76.3±11.8 and 3.4±2.3 µg.g⁻¹ respectively than in full sun cocoa beans. Regarding aldehydes, agroforestry dry fermented cocoa beans recorded 65.32±3.32 µg.g-1 of methyl butanol while full sun cocoa beans recorded 26.3±9.7 µg.g⁻¹. 2-Phenylethanol was found at the concentrations of 75.4±54.3 and 22.7±8.2 µg.g⁻¹. 2-Phenylethanol was found at the concentrations of 75.4±54.3 and 22.7±8.2 µg.g⁻¹ in agroforestry and full sun cocoa beans samples, respectively. 2-Phenylethanol was found at the concentrations of 75.4±54.3 and 22.7±8.2 kg.g-1 µg.g-1 in agroforestry and full sun cocoa beans samples, respectively. 2-Phenylethanol was found at the concentrations of 75.4±54.3 and 22.7±8.2 kg.g-1 µg.g-1 in agroforestry and full sun cocoa beans samples, respectively.µg.g-1 in agroforestry and full sun cocoa beans samples, respectively. Among pyrazines, tetramethyl pyrazine was found at highest concentration in dry fermentation produced in Guiberoua’s area. The concentration of specific acetate esters including ethyl acetate and phenylethyl acetate can be favoured by fermentation techniques like turning the beans [55]. The richness of Gubéroua cocoa in various relevant aroma compounds may be due to the flavors produced by yeast species such as Saccharomyces cerevisiae, Pichia kudriavzevii and Hanseniaspora opuntia in the fermentation of cocoa carried out in this region [28,56]. Furthermore, Crafack et al. [54] have previously demonstrated that several flavor compounds are produced primarily by yeast and bacteria during the fermentation process, which converts amino acids and higher alcohols into various aroma precursors. According to Ho et al. [46] and Ziegleder [57], the aroma profile of chocolate is enhanced by significant increases in esters concentration during spontaneous fermentation. The distinct floral and fruity notes that distinguish premium chocolate from bulk varieties are due to other key compounds like linalool (found in fine cocoa varieties) and phenylethyl acetate [12]. Linalool is the essential aroma compound that forms during cocoa fermentation, as Ziegleder [58] has reported for a considerable amount of time. Our findings that tetramethyl pyrazine is the principal compound of the pralines class were in accordance with those previously obtained [59]. Although several works reported that pyrazine group is more common in well-fermented roasted beans, these metabolic products of Bacillus species (B. subtilis or B. megaterium) are formed at the end of cocoa fermentation [40,41,60,61]. Each flavor compound could contribute to the final aromatic quality of dry fermented cocoa beans. For example, pyrazine compounds have cocoa, chocolate, walnut, popcorn, coconut, candies, fruity notes [42]. However, certain aroma compounds were more concentrated in agroforestry cocoa beans than those of full sun from the Guiberoua region. The agroforestry system has yet to be shown to have an impact on the formation of crucial aroma compounds in dry-fermented cocoa beans.

3.7. Effects of Agroforestry on the Flavor Compound Profiles of Chocolates Produced from Dry Fermented Cocoa Within Each Producing Geographical Origin.

Figure 5 and Figure 6 indicate the PCA biplots, which reveal the effects of agroforestry on the flavor profiles of chocolate between and within production areas tested, respectively. The primary results about between cocoa producing areas revealed that only the flavor profile of chocolate produced from agroforestry cocoa beans was discriminated from those of chocolate derived from full sun cocoa beans in only the Méagui region. However, our findings about each cocoa producing area showed that the flavor compound profile of chocolate derived from agroforestry cocoa beans sourced from the areas of Agnibilékrou (Figure 6-B) and Méagui (Figure 6-E) was discriminated from those of chocolate samples derived from full sun cocoas corresponding. The agroforestry system has shown no significant effect on the flavor profiles of the chocolates issued from other cocoa-producing areas consisting of Adzopé (Figure 6-A), Divo (Figure 6-C) and Guiberoua (Figure 6-D) regions. The flavor profiles of agroforestry chocolate from Agnibilékrou region recorded hexanal, ethyl acetate, ethanol, sec butyl acetate, acetoin and butane-2,3-dione while those made from full sun dry fermented cocoa beans showed 2,3-butanediol, butyl acetate, 3 methyl butanol, Pentan-2-ol, benzyl acetate, ethyl octonate, ethyl phenyl acetate, 3,5 dimethyl-2-ethyl pyrazine, acetophenone, 2-phenyl acetate, 2,3,5-trimethyl-6-ethyl pyrazine. The chocolate made from agroforestry cocoa beans produced from Méagui contained more aroma compounds (23) than those derived from full sun cocoa. The flavor profile of this agroforestry chocolate included several key aroma compounds consisting of benzyl alcohol, isobutanol, benzylacetate, ethylphenylacetate, 2-phenyl butanol, benzaldehyde, 2-phenyl ethanol, isobutyl acetate, 3-methyl butanoic acid. We may remind that the flavor profiles of agroforestry dry fermented cocoa beans from Méagui and Agnibilekrou consist of isobutyric and isomeric acids. Isobutyric acid is present in cocoa beans, where it is formed during the fermentation and drying processes from the decomposition of sugars and other compounds [62]. Although it contributes to the overall flavor profile of the beans, its concentration may vary based on factors such as fermentation time, bean variety, and subsequent processing like roasting [63]. However, isovaleric acid is present in cocoa beans, where it is a volatile organic acid that develops during fermentation. It is produced by microorganisms and contributes to the acidification of the beans, though it is considered an undesirable compound when present in high concentrations, as it can produce a "rancid" smell. Although the sensory quality of chocolate broadly depends on the volatile compound profile resulting from microbial metabolism during cocoa beans fermentation [64], the discrimination of the flavor profiles of chocolates derived from agroforestry cocoa beans produced in the Agnibilékrou and Méagui regions could be due to the presence of either one, the other, or even both acids (isovaleric and isobutyric acids).

3.9. Effects of Agroforestry on the Sensory Perception of Chocolate Samples

Sixteen descriptive attributes, namely intensity of odor, acid, bitter, sweet, astringent, cocoa aroma, fresh fruit, dried fruit, floral, spicy, toasted, woody, plant, alcoholic aroma, animal, and global quality, were evaluated. Average of the score attributed for each sensory property to the finished chocolates from each fermented cocoa sample was then calculated. Figure 7 and Figure 8A-B show that all end-chocolate samples present the same score for almost all descriptive attributes regardless both of cropping system and producing geographical origin. Neither agroforestry nor the cocoa production area has influenced the sensory perception of manufactured chocolates.

4. Conclusions

This study investigated the effect of agroforestry on the flavor profiles both of cocoa beans and of the chocolate’s sensory perception. Agroforestry significantly influenced the native flavor profiles of raw cocoa beans regardless of the producing geographical origin. After transformation (fermentation and roasting), the difference in flavor compound profiles is not as significant as it was in fresh cocoa beans, because of yeasts and thermal treatments. Furthermore, agroforestry influenced the flavor profile of dry fermented cocoa beans and end-chocolate in function of the producing area. Agroforestry has shown no effect on the sensory perception of chocolate. Agroforestry can therefore be promoted in cocoa cultivation around the planet in order to reduce the impact of deforestation and ensure the sustainability of cocoa. The outcome of this study will be to evaluate the effect of agroforestry with the same density of trees of even height on the organic qualities of chocolate by integrating soil quality.