Article pubs.acs.org/JAFC
Investigations on the Aroma of Cocoa Pulp (Theobroma cacao L.) and Its Influence on the Odor of Fermented Cocoa Beans Irene Chetschik,* Markus Kneubühl, Karin Chatelain, Ansgar Schlüter, Konrad Bernath, and Tilo Hühn Institut für Lebensmittel und Getränkeinnovation ILGI, Zürcher Hochschule für Angewandte Wissenschaften, Grüental, 8820 Wädenswil, Switzerland ABSTRACT: The odor-active constituents of cocoa pulp have been analyzed by aroma extract dilution analysis (AEDA) for the first time. Pulps of three different cocoa varieties have been investigated. The variety CCN51 showed low flavor intensities, in terms of flavor dilution (FD) factors, in comparison to varieties FSV41 and UF564, for which floral and fruity notes were detected in higher intensities. To gain first insights on a molecular level of how the cocoa pulp odorants affected the odor quality of cocoa beans during fermentation, quantitative measurements of selected aroma compounds were conducted in pulp and bean at different time points of the fermentation. The results showed significantly higher concentrations of 2-phenylethanol and 3methylbutyl acetate in pulp than in the bean during the different time steps of the fermentation, whereas the reverse could be observed for the odorants linalool and 2-methoxyphenol. The findings of this study constitute a basis for further investigations on the aroma formation of cocoa during fermentation. KEYWORDS: cocoa pulp odorants, Theobroma cacao L., aroma extract dilution analysis, aroma formation in cocoa, cocoa fermentation
■
INTRODUCTION The unique, mouthwatering flavor of cocoa is appreciated by people all around the world and is one of the most important reasons for the consumption of chocolate, the product from the seeds of the cocoa tree (Theobroma cacao L.) As defined by the International Cocoa Organization, there are two main categories distinguishing cocoa according to their flavor properties on the world market, namely, fine or flavor and bulk cocoa.1 Fine or flavor cocoa is exclusively produced from Criollo, Trinitario, and National types, which are known for their extraordinary flavor properties, in particular for their pronounced fruity, floral aroma attributes, whereas bulk cocoa, a cocoa with average flavor properties, is mainly produced from Forastero cocoa.2 By comparison of the market shares of bulk and fine or flavor cocoa on the world cocoa market, bulk cocoa, which is used to produce cocoa products such as cocoa mass, cocoa powder, and chocolate, is, with its market share of 90%, significantly more predominant than fine and flavor cocoa, with an estimated market share of only 5−10%, and thus, is mainly used for the production of high-quality chocolate.2,3 Before the cocoa seeds can be used as raw material for chocolate products, many different processing steps, such as fermentation, drying, and roasting, have to be conducted to transform the unpleasant aroma of raw cocoa seeds into the characteristic mouthwatering cocoa aroma.4 Thereby, the fermentation process is considered as the most important factor for the flavor development of cocoa beans,5 and it is known that raw or unfermented cocoa beans elicit no characteristic cocoa aroma when roasted.6 The cocoa pulp, which surrounds the raw unfermented cocoa beans, plays an important role during fermentation. As previously summarized,7 the cocoa pulp is first bioconverted by yeasts (anaerobic phase) and in a later stage by lactic acid bacteria (aerobic phase). In this way, organic acids such as acetic and lactic acid © XXXX American Chemical Society
are formed, which permeate the cocoa seed tissue, whereby a rise of the temperature in the fermentation mass takes place, leading to the degradation of the cocoa tissue by decomposition of proteins and carbohydrates by the cocoa enzymes to flavor precursors such as reducing sugars and amino acids.5,8 Besides, the aromatic cocoa pulp is assumed to have an impact on the aroma development of cocoa seeds during the fermentation process by the possible migration of aroma compounds from the pulp into the seed tissue and is regarded as a reservoir for fine or flavor cocoa notes.9 Although a large number of research studies can be found elucidating the key flavor compounds in cocoa beans, depending on varieties and processing,10−12 only a few studies investigated the aroma constitution of cocoa pulp.13,14 In the latter studies, quantitative and qualitative differences between the volatile constituents of various pulp varieties could be detected by gas chromatography−mass spectrometry (GC-MS) and were assumed to be responsible for the dissimilarities in the sensory profiles of the different cultivars, in particular between pulps of bulk and fine or flavor cocoa. For this reason, the hypothesis that fine or flavor cocoa notes are linked to the flavor characteristic of the pulp9 could be consolidated. However, the contribution of identified volatiles to the overall aroma impression was not verified by gas chromatography−olfactometry (GC-O) in combination with dilution to odor threshold techniques, such as aroma extract dilution analysis (AEDA).15 Therefore, the first objective of this investigation was to screen for the key Special Issue: 11th Wartburg Symposium on Flavor Chemistry and Biology Received: Revised: Accepted: Published: A
November 8, 2016 March 9, 2017 March 20, 2017 March 20, 2017 DOI: 10.1021/acs.jafc.6b05008 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
conditions were chosen to be as close as possible to the conditions of the fermentation at the place of origin. During the first 48 h of the fermentation, anaerobic conditions were maintained, and the temperature was adjusted after 24 h to 32 °C and after 48 h to 38 °C. After 48 h, the aerobic phase was started by removing the lid of the bottle and maintaining the contact of the fermentation material with oxygen by the gentle stirring of the mixture on a magnetic stirrer. The temperature of the fermentation mixture was adjusted (at 72 h to 45 °C and at 96 h to 50 °C). Because no real flora of the fermentation at the place of origin was available for application in the laboratoryscale fermentation, a common red wine yeast (Saccharomyces cerevisiae) was solely used as fermentation culture for the following reasons: yeasts are known to be the most important and predominant microorganisms in the fermentation of cocoa,16 whereas lactic acid bacteria and acetic acid bacteria were shown to be less important as yeasts for aroma formation during fermentations.17,18 In addition to that, red wine yeasts are known to operate well at higher fermentation temperatures,19 and good results could be achieved in fermenting cocoa pulp by this microorganism in preliminary experiments (data not shown). This way, the fresh cocoa pulp from the variety FSV41 (42 g) was spiked with 15 mg of yeast (red wine yeast, Danstar Ferment AG, Fredericia, Denmark) for the laboratory-scale fermentation. Before the spiking, the yeast was washed with 15 mL of sodium chloride solution (1%), centrifuged (4000 rpm for 5 min), and rehydrated with 15 mL of tap water. For the vitalization of the yeast, 1.5 mL of aqueous glucose solution (1.33 g/mL) was added at a temperature of 27 °C. After 60 min, the yeast−glucose solution and the cocoa pulp were added into a durex bottle (75 mL). During fermentation, samples of the pulp were taken at time points 24, 48, and 96 h, after a gentle homogenization of the fermentation material by stirring. The samples were spiked with internal standard mixture (labeled and unlabeled compounds) and stored at −20 °C until the analysis was to be conducted. Methods. Isolation of the Volatiles for GC-O. For the screening on the aroma-active compounds, 50 g of material retrieved from the different pulps was extracted with 100 mL of dichloromethane by vigorous stirring at room temperature. To separate the volatiles from the nonvolatile material, the pulp−dichloromethane mixtures were directly applied into the SAFE distillation,20 the head and the legs of the apparatus were thermostated at a temperature of 40 °C, and the apparatus was connected to a vacuum pump generating a vacuum of 5 × 10−6 to 5 × 10−5 Pa. The obtained distillates were dried over anhydrous sodium sulfate and concentrated on a Vigreux column to a final volume of 300 μL. The obtained extracts were analyzed by comparative aroma extract dilution analysis by GC-O. Workup Procedure for the Quantitative Analyses in Bean and Pulp during the Fermentation Process. For the quantitative screening of selected flavor compounds (see Table 1), which constituted the main key aroma compounds in cocoa10,11 and could be also confirmed
aroma compounds in three different cocoa pulp varieties by GC-O, to assess their impact on the overall aroma impression by the performance of AEDA, and to verify the pulp key odorants by identification experiments with GC-MS, retention index data, and comparison with reference substances. The second aim of the study was to provide first insights if odorants responsible for the fine or flavor notes might indeed enter into the seed tissue during the fermentation process and therefore can be considered as influential on the overall odor quality of fermented cocoa beans by quantitative measurements of selected key aroma compounds of pulp and bean of one variety during the fermentation process.
■
MATERIALS AND METHODS
Cocoa Pulps and Cocoa Seeds. Cocoa pulps of different varieties and origins (FSV41 from Colombia, Region Santander, and UF654 and CCN51, both from Costa Rica; confirmation of the varieties has been conducted on the basis of the morphology of the analyzed material) were extracted from the cocoa seeds immediately after harvesting and opening of the pods at the place of origin. This was done by first opening the pods with a sharp cutlass, then dislodging the beans and pulp from the shell, and removing the cocoa pulp surrounding the single beans by rubbing the pulp bean units on a stainless steel sieve with a mesh size of 8 mm. To avoid contamination of the pulp with microorganisms from the surrounding to the utmost extent, the removed pulps were immediately blended with saturated calcium chloride solution (1:1), then sealed in odorless plastic bags, and immediately frozen to −20 °C to stabilize the investigation material. The samples were then transported in a frozen state from the country of origin to Switzerland. For the investigation on the aroma transformation in bean and pulp during fermentation, pulp and cocoa seeds of the variety FSV41 sampled at different time points of fermentation were more closely investigated. For the time point “0 h” of the experiment, fresh pulp and seeds were collected directly after the harvest. The pulp was treated and packed as described above, whereas the seeds were packed and sealed in plastic bags and frozen immediately to −20 °C. Additionally, the cocoa seeds were submitted to a single-variety fermentation at the place of origin. For the fermentation, the pulp and seeds of FSV41 (21 kg fermentation mass) were placed in a wooden fermentation container and covered by plastic and wooden panels for the first 48 h. The fermentation of the material started spontaneously by the natural flora (no additional fermentation cultures were added) and continued for 96 h, during which the material was aerated every 24 h with a wooden shovel. The temperatures measured during the fermentation were 26.8 °C at the starting point, 31.8 °C at 24 h, 37.5 °C at 48 h, and 50.4 °C at 96 h. The end point of the fermentation (96 h) was determined by means of the cut test of the beans, which was performed when the beans in the fermentation mass showed a distinct brown color. Bean samples were taken throughout the fermentation process after 24, 48, and 96 h, respectively. The seed sample “0 h” (= fresh sample) was directly taken after pod opening (as described above). After the sampling, the seeds were sealed in plastic bags, immediately frozen to −20 °C, and shipped from the country of origin to Switzerland in a frozen state (−20 °C). Because the sampling of the pulp during fermentation (especially at last time points: 48 and 72 h) was not possible due to the liquefaction of the pulp and its drainage out of the wooden fermentation container, a laboratory-scale model fermentation of the fresh pulp (collected at the place of origin) was performed to get insights into the aroma transformation in the pulp during different time points of the fermentation. For this reason, the fresh pulp sample FSV41 was collected after pod opening as described above, directly frozen to −20 °C (without the addition of saturated calcium chloride solution) after sealing in plastic bags, and shipped to our laboratory in Switzerland, where it was treated as described in the section below. Laboratory-Scale Fermentation. For the laboratory-scale model fermentation of the FSV41 cocoa pulp sample, the fermentation
Table 1. Selected Ions (m/z) of Analytes and Standards and their Response Factors (Rf) Used for Quantitation ion (m/z)a odorant acetic acid 2-phenylethanol linalool 2-methoxyphenol 3-methylbutanoic acid 3-methylbutyl acetatec
isotope label 13
[ C2] [2H5] [2H5] [2H3] [2H3]
analyte
standard
Rfb
60 122 121 124 60 87
62 127 126 127 63 70d
1.00 0.91 1.26 1.14 1.80 0.88
a
Ion used for quantitation in electron ionization EI mode. bResponse factor (Rf) was determined by analyzing mixtures of known amounts of analyte and standard cCompound was quantified via the unlabeled standard pentylacetate dIon of pentylacetate selected for the quantification of 3-methylbutyl acetate. B
DOI: 10.1021/acs.jafc.6b05008 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry Table 2. Odor-Active Compounds Identified in Aroma Distillates Isolated from Cocoa Pulps of Different Varieties FD factora
retention index on no.
b
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37
odorant
c
2,3-butanedione methyl 2-methylbutanoate hexanal unknown 3-methylbutyl acetate octanalf 1-octen-3-onef unknown (Z)-1,5-octadien-3-onef acetic acid 3-(methylthio)propanal (E,E)-2,4-heptadienalf (Z)-nonenalg 2-isobutyl-3-methoxypyrazinef linalool 2-acetylpyrazinef phenylacetaldehyde 2-acetylthiazolef (E,Z)-2,4-nonadienalg 2- and 3-methylbutanoic acid unknown (E,E)-2,4-nonadienalf (E)-2-undecenalf ethylphenyl acetate (E,E)-2,4-decadienalf 2-methoxyphenol ethyl 3-phenylpropanoatef 2-phenylethanol trans-4,5-epoxy-(E)-2-decenalf γ-nonalactonef Furaneol 4-methylphenolf 4-vinyl-2-methoxyphenolf sotolonef unknown unknown phenylacetic acid
odor quality
d
FFAP
buttery fruity green sulfury fruity citrus-like mushroom-like nutty geranium-like, green pungent, sour cooked potato citrus-like cardboard-like earthy, bell-pepper citrus-like, floral nutty flowery, honey nutty, roasty cardboard, fatty rancid fruity, flowery fatty, green tallowy, sweet sweet tallowy, fatty smoky fruity flowery, sweet metallic coconut-like caramel-like phenolic, animalic smoky smoky smoky sweet, coconut-like sweet
1010 1017 1086 1109 1114 1278 1288 1354 1361 1430 1437 1473 1493 1511 1537 1605 1623 1633 1640 1654 1682 1695 1713 1727 1814 1848 1872 1898 2014 2019 2048 2081 2195 2203 2347 2482 2590
DB-5 775 809 880 1005 980 982 600 886 1010 1147 1186 1092 1029 1037 1194 860 1255
1330 1090 1113 1391 1359 2033 1077 1077 1196
1281
FSV41
CCN51
UF654
2 2 1 4 2 2 8
