




































In ternationa l
Scholars
Journa ls

 

African Journal of Agricultural Marketing ISSN 2375-1061 Vol. 4 (12), pp. 001-006, December, 2016. Available online 
at www.internationalscholarsjournals.org © International Scholars Journals 

 

Author(s)                                           Author(s) retain the copyright of this article. 
 

 

Full Length Research Paper 

 

Impact of cocoa processing technologies in free fatty 

acids formation in stored raw cocoa beans 
 

Simplice Tagro Guehi 1*, Michael Dingkuhn2, Emile Cros3, Gérard Fourny3, 
Robert Ratomahenina4, Guy Moulin4 and Anne Clement Vidal2  

1
Laboratoire de Biochimie et de Technologie Alimentaire. UFR Sciences et Technologies des Aliments. Université 

d’Abobo-Adjamé. 02 Bp 801 Abidjan 02. Côte d’Ivoire. 
2
CIRAD-AMIS, Equipe Ecotrop, TA 40/01, 34398 Montpellier Cedex 5, France. 

3
CIRAD-CP/Programme Cacao, Laboratoire de Chimie-Technologie du cacao, TA 80/16, 34398 Montpellier Cedex 5, 

France. 
4
INRA-Montpellier, Laboratoire de microbiologie industrielle et de génétique des microorganismes, 2 Place Viala, 34000  

Montpellier, France. 
 

Accepted 18 July,  2016 
 

The quality of raw cocoa beans depends widely on their free fatty acids (FFA) content. High FFA 
content is a serious quality defect and reduces the technical and economic value of the cocoa beans. 
The work investigates the influence of cocoa processing technologies on FFA formation during storage 
of raw cocoa beans. Different samples of ferment dried cocoa beans purchased from Cote d’Ivoire were 
stored and analysed for FFA content. Very low FFA contents were found in whole healthy cocoa beans 
generally complied with UE standards (1.75% oleic acid equivalent) throughout storage while high FFA 
content was found in poor quality and broken healthy beans. The formation of FFA did not depend on 
the genotype or on cocoa post-harvest processing technologies. However, high and increasing FFA 
contents were observed in defective cocoa beans and could be attributed probably to the activity of 
microflora which in turn were associated with initial quality and loss of physical integrity of the cocoa 
beans. 

 
Key words: Raw cocoa beans, processing technologies, free fatty acids content. 

 
INTRODUCTION 

 
Cocoa beans are actually seed from the fruit of 
Theobroma cacao tree. Cocoa butter, the naturally fat, 
contained in cocoa beans is the only continuous phase of 
chocolate. It is responsible for dispersion of the other 
constituents of chocolate properties during storage, 
handling and tasting (Nickless, 1996). Cocoa butter was 
currently reported to be the main vegetable fat used in 
the chocolate manufactures because of its rheological, 
textural and chemical characteristics such as triglycerides 
fatty acids composition, (Awua, 2002; Lipp and Anklam, 
1998; Whithefield, 2005). Its hardness depends on the 
saturated and unsaturated fatty acid contents bound in 
triglycerides, and on free fatty acids (FFA) content. The  
 
 
 
*Corresponding author. E-mail: g_tagro@hotmail.com. Tel: +  
225 66 11 23 44. Fax : + 225 20 37 74 48. 

 
 
 

 
general opinion is that higher FFA content leads to a 
decrease in hardness of cocoa butter (Pontillon, 1998) 
and must be considered as a raw cocoa commercial 
value reducing factor both for producers and chocolate 
manufacturers. For reasons of quality therefore, the 
directive 73/241/EEC (EEC, 1973) limits the maximum 
FFA content to 1.75% oleic acid equivalent in cocoa 
butter. Yet, for several years, from 15 to 20% of annual 
Ivorian cocoa production is considered to have excessive 
FFA contents (exporters' statistics) recurrently and 
seasonally, notably at the end of the main season.  

FFA are carboxylic acids released from triglycerides 
(Selamatet al., 1996) through the effect of a lipase (E.C. 
3.1.1.3) or an oxidation. The risks of oxidation are 
negligible in cocoa butter due to its low unsaturated fatty 
acid content (Withefiled, 2005) and high content of 
polyphenols, natural antioxidants, in cocoa beans 
(Nickless,1996). To date, although no lipase activity has 

file:///C:\Users\user\Documents\REPUBLICATION\AGRICULTURAL%20SCIENCES\AppData\Local\Temp\www.internationalscholarsjournals.org


 
 
 

 

yet been detected in fermented and dried cocoa beans to 
our knowledge, the role of a lipase is strongly suspected. 
Numerous studies have shown plant lipase activity is 
mostly detectable during seed germination (Wanasun-
dara et al., 2001), except in some oil crops such as the 
castor-oil plant (Ory et al., 1962), nigella kernel (Mert et 
al., 1995) and rice (Raghavendra and Prakash, 2002). 
For Fowler (1999), high FFA contents in cocoa beans 
might result from black beans originating from rotten pods 
or germinated beans. Likewise, microflora, particularly 
moulds, can cause similar problems during storage (Hiol, 
1999). Indeed, Wood and Lass (1985) and Pontillon 
(1998) suggested that FFA occurrence in stored cocoa 
beans is linked to the action of microbial lipases. 
Therefore, this study aims to determine the impact of 
cocoa processing technologies on FFA formation in raw 
cocoa beans and if microflora of the cocoa beans are 
implied in FFA formation. 
 

 
MATERIALS AND PROCESSING 
 
Cocoa beans used in this study were originated healthy from region 
of Soubré in Côte d’Ivoire. A total of 16 samples (25 pods each) 
were opened and beans of each protocol were fermented using the 
micro-fermentation technique in a wooden box measuring 65 x 50 x 
50 cm (turned manually after 48 and 96 h). Cocoa beans were sun 
dried on plastic tarpaulins for one week, reducing bean moisture 
content from 60 to 8%. All samples were stored in a climatic cabinet 
(Firlabo, SB-BVEHF type) . Temperature and relative humidity were 
fixed respectively at 27°C and 75% for 6 months. For storage, 50 g 
of cocoa beans were individually putted in bags to be removed at 
each sampling time, 0, 2, 3, 4 and 5 months later. FFA content was 

measured on carefully shelled, manually and finely ground beans ( 
500 µm).  

Effect of cocoa beans processing technologies on FFA’s forma-
tion was studied. Genetic origin’s effect was studied using beans 

from three cultivars ripe pods: Amelonado, Ivorian 1
st

 generation of 
hybrids (Amelonado x West African Trinitario) and open pollinated 
progenies. Influence of pod degree ripeness was studied using 

beans from pre-ripe, ripe and over-ripe pods of Ivorian 1
st

 genera-
tion of hybrids. Pods were opened 5 days after harvesting. Beans 
from 1st generation hybrids pods harvested the same day were 
used for the impact of the time lapse between pods harvesting and 
opening. Pods were divided into three parts and were opened 
respectively the day of harvesting, 5 and 9 days after harvesting. 
Ripe pods from open pollinated progenies were used for study of 
fermentation duration’s effect. Beans samples were taken in the 
middle of the cocoa beans mass each day throughout fermentation 
period and then were sun dried.  

Effect of raw cocoa beans quality on FFA formation was studied 
using 1200 g of uncontrolled cocoa samples of three different quail-
ties: healthy beans which are brown and compact with no visible 
defects, clustered beans and black beans originated from Abidjan 
harbour. Clustered beans are the agglomerates of 2 to several 
beans. Each sample was divided into 4 fractions: beans of first 2 
fractions were individually and roughly broken and those of second 
2 fraction remained whole. And then both sub-samples were stored 
(27°C, RH 75%). FFA content was measured on 50 g sampled from 
each fraction every two weeks.  

About 1800 g of black beans samples were used in order to 
determine influence of decontamination of the cocoa beans and 
beans physical integrity in the FFA’s formation in raw cocoa beans. 

The samples were divided into 6 equal portions: 2 untreated cocoa 

  
  

 
 

 
beans of first fraction were stored as previously described, the 
beans of the 2 others parts are individually and roughly broken 
(control) and cocoa beans of last 2 fractions were broken and then 
individually and carefully decontaminated by soaking in a 5% (v/w) 
hydrogen peroxide solution (Sigma) for 3 hours at 28°C. After 
sterilely drying at 27°C during 3 days about 40 g of each decon-
taminated cocoa beans were individually sealed hermetically into 
sterile flasks prior to storage for 12 weeks as previously described. 
FFA content was measured every two weeks on samples from one 
flask per experimental protocol and from over corresponding sub-
samples. 
 

 
Physicochemical analyses 
 
About 15 g of cocoa beans were carefully shelled manually. Cocoa 
nibs were frozen in liquid nitrogen before finely grinding in a 
kitchen- scale coffee grinding (Moulinex, France) to the smallest 

particle (size  500 µm). Ten grams of cocoa powder were putted in 
Whatman cartridge and soaked in 350 ml of petroleum ether 
(Prolabo Normapur, type 40-60°C) for one night. Cocoa butter was 
extracted on a Soxhlet apparatus for 8 h. After eliminating of solvent 
in a rotary evaporator, FFA contents were quantified by the official 
method 42-1993 (IOCCC, 1996). About 5 g of extracted cocoa 
butter was weighed (W) and dissolved in 50 ml of a previously hot 
petroleum ether/absolute ethanol mixture (1:1, v/v) neutralized by 
adding phenolphthalein. Titration carried out with 0.1N alcoholic 
KOH solution used (V) was noted. FFA (% oleic acid) was 
calculated as follows: 
 

FFA %  oleic  acid   
28 .2  V  N   

 

W  

 
 

 

Statistical analysis 
 
Data represent the means of triplicate determinations. Analysis of 
variance were conducted using JMP software Version 5 (SAS 
Institute, 2002) with mean separation performed by the LSD test 
(P<0.05). Quadratic regression function at 0.05 confidence intervals 
was conducted using SigmaPlot 8.0 software. 
 

 

RESULTS AND DISCUSSION 

 

Effect of cocoa processing technologies on the FFA 

content 
 
FFA content in whole healthy beans was low and ranged 
from 0.2 to 0.8%, irrespective of the genotype (Figure 1). 
FFA contents are not significantly different whatever the 
genotype of cocoa. In the similar FFA content of pods 
ripeness degree (results not shown) or pods opening 
delay are around the same values (Figure 2). 

Slighter increasing FFA depending on the fermentation 
duration indicates a difference (LSD test, P<0.05) within 
fermented cocoa beans. Initial and final FFA contents in 
cocoa beans fermented during over 3 days are higher 
than those in cocoa beans fermented below 3 days. 
Cocoa beans fermentation duration seemed to be critical 
to increase the chances for FFA formation. Furthermore 
after 5 months' storage FFA contents remained below 



 
 
 
 
 
 
 

 

F
re

e
 F

a
tt

y
 a

c
id

s
 c

o
n

te
n

t 
(%

) 

 
 
 
 
 
0,8  
 
 

 

0,6  
 
 

 

0,4       

    Amelonado   
    Free descendances  

0,2    Hybrids   

0,0       
0 1 2 3 4 5 6   

Storage time (months) 
 

Figure 1. Effect of cocoa genotype on FFA content during storage at 27°C, 75% 

HR for up to 5 months. Bars represent the standard error of the mean (n=3). 
 
 
 
 
 

 

F
re

e
 F

a
tt

y
 A

c
id

s
 c

o
n

te
n

t 
(%

) 

 
 

 

0,7  

 

0,6  

 

0,5  

 

0,4  

 

0,3  
 

2 days after harvesting  0,2 
 5 days after harvesting  

9 days after harvesting  
0,1  

 

0,0  
0 1 2 3 4 5 6 

 
Storage time (months) 

  
Figure 2. Effect of pod opening delay on FFA content in cocoa beans during storage at 

27°C, 75% HR for up to 5 months. Bars represent the standard error of the mean (n=3). 
 

 

1.75% (Figure 3). The small changes observed in FFA 
content in controlled cocoa samples indicate that any 
appropriate cocoa processing technology did not have an 
appreciable impact on FFA formation during correct 
storage. 

 
 

 

Effect of cocoa beans quality, physical integrity and 

decontamination on FFA content 
 
Initial FFA contents in uncontrolled cocoa beans sampled 

according their quality at Abidjan harbour were over 12- 



  
 
 
 
 
 
 

 

F
re

e
 F

a
tt

y
 A

c
id

s
 c

o
n

te
n

t 
(%

) 

 
 

 
0,8  
 
 

 
0,6 
 
 

 

0,4       

    1 day of fermentation  
    2 days of fermentation  

0,2    3 days of fermentation  

    4 days of fermentation  

    5 days of fermentation  

    6 days of fermentation  

0,0       
0 1 2 3 4 5 6 

 
Storage time (months) 

  
Figure 3. Effect of cocoa beans fermentation duration on FFA content during storage at 

27°C, 75% HR for up to 5 months. Bars represent the standard error of the mean (n=3). 
 
 
 
 

 

F
re

e 
F

at
ty

 A
ci

d
s 

co
n

te
n

t 
(%

 w
/w

) 

 
 

 
 

20  

18  

16 Whole clustered beans 
 Artifically brocken clustered beans 

14   
 
12 

 
10 

 
8 

 
6 

 
4  

0 2 4 6 8 10 12 14 
 

Storage time (weeks) 
  

Figure 4. Influence of artificially broken clustered beans on FFA content during storage at 

27°C, 75% HR for up to 2 months. Bars represent the standard error of the mean (n=3). 
 

 

fold higher in both clustered beans (6.23%) and black 
beans (7.48%) than those in healthy and intact beans 
(0.48%). The low FFA content in whole healthy beans 
increased from 0.48 to 0.78% and did not exhibit an 
appreciable change during over 12 weeks' storage 
(results not shown). Considerable increasing FFA con-
tents were observed in poor quality cocoa beans such as 
clustered beans where it increased from 6.23% to ca 

 
 

 

11% (Figure 4). No variation of FFA content was noted in 
black beans where FFA content remained around 10% 
(results not shown).  

Bean clustering could have been caused either by 
harvesting of unripe pods, or by poor separation of the 
beans and placentas during cocoa pods opening. So 
beans would be clustered in placentas during and after 
fermentation. Barel (1998) have mentioned that following 



 
 
 
 
 
 

 

F
re

e 
fa

tt
y

 a
ci

d
 c

o
n

te
n

t 
(%

 w
/w

) 

 
 
 

 
25   

Artificially broken and untreated beans  
 
20  
 

 

15  
 

 

10  
 

 

5 
Artificially broken and decontaminated beans 

 

 
 

0  
   

0 2 4 6 8 10 12 
 

Storage time (weeks) 
  

Figure 5. Variation of FFA content in decontaminated artificially broken black beans during 

storage. Solid lines indicate quadratic regression functions and broken lines indicate 0.05 

confidence intervals for each function. 
 

 

inadequate drying due to a surface crust clustered cocoa 
beans show always high moisture content which favoured 
moulds growth. According to Poisson and Cahagnier 
(1979) and Raghavendra and Prakash (2002) foodstuffs 
with high moisture content were easily attacked by 
moulds. These moulds could produce lipase (Wood and 
Lass, 1985) which in contact with cocoa butter of broken 
cocoa nibs released FFA from triglycerides. Black beans 
were probably originating from overripe pods infected by 
Phytophthora sp. Fungal diseases could create optimal 
conditions for the development of moulds (Renaud, 
1954), which in turn would lead to FFA formation. 
However, variable FFA content of black bean sample was 
probably due high variation among individual beans and 
insufficient sample size. Given the variation of FFA con-
tent due probably to a highly heterogeneity of samples, it 
was not possible to characterize changes in FFA contents 
in storage conditions used. High FFA contents were due 
to poor quality of cocoa raw materials quality as 
previously demonstrated (Guénot et al., 1976).  

Slighter increasing of lower FFA contents in whole 
healthy beans than that in defective beans confirms that 
storage time did not affect FFA formation if processing 
technologies were previously appropriate. The greater the 
initial FFA content of the raw cocoa beans, or the lower 
the quality of the beans, the greater is this increase in 
FFA. And when cocoa beans were artificially broken, their 
FFA content increased strongly whatever cocoa beans 
initial quality. FFA content of decontaminated artificially 
broken beans remained unchanged and well below the 
considerably increasing FFA content in the 

 
 

 

untreated cocoa beans. Using the quadratic regression 
functions at 0.05 confidence intervals indicates that the 
variations over storage time in FFA content of untreated 
cocoa beans sample and decontaminated sample are 
significantly different (Figure 5).  

As decontamination prior to storage under sterile condi-
tions stopped FFA formation in broken black beans only 
lipase activity of external origin can thus be considered. 
This conclusion was supported by many studies which 
demonstrated that plants in which lipases are detected 
during dormancy are rare. This is the case with the 
castor- oil plant (Ory et al., 1962) and nigella (Mert et al., 
1995). Indeed, enzyme lipase is responsible for lipid 
breakdown in plants to produce the energy required for 
embryo growth (Imeson et al., 1993). Yet, during cocoa  
beans fermentation, the rise in temperature (50°C) and 
diffusion of acid products through the cotyledons due to 
partial cell walls lyses, hence death of germ and prevent 
germination (Barel, 1998). In addition fermentation pro-
cess induced contact between proteins and polyphenolic 
compounds hence inactivation of enzymatic proteins. 
Indeed many polyphenolic compounds in plants were 
already described as enzyme activities inhibition factor 
(Hagermann, 1992; Kallithraka et al., 2001; Misnawi et 
al., 2002). So cocoa beans endogenous enzymes could 
not be active in complete ferment dried cocoa beans 
(Yusep et al., 2002). Cocoa beans of poor quality 
deteriorate much faster than cocoa beans of hight quality. 
Circumstantial evidence was presented that lipolytic 
activity leading to the FFA formation was due to microbes 
associated with other factors such as the quality, the phy- 



 
 
 

 

sical condition, and the storage conditions of cocoa 
beans (Guénot et al., 1976, Oyeniran, 1979; 1980) and 
not to endogenous plant lipases by decontaminating part 
of the cocoa beans with hydrogen peroxide. Similar 
results were obtained by Hiol (1999) in previous study on 
formation of FFA in palm fruits caused by moulds. Thus 
cultivation and identification of microbes responsible of 
FFA formation in raw cocoa beans was not elucidated in 
this study, yet the relatively high FFA content exhibited by 
the poor quality beans and no change observed in FFA 
content of decontaminated beans were good evidence to 
support their potential lipase-producing abilities and their 
action on cocoa butter. 
 

 

ACKNOWLEDGEMENTS 

 

This research was supported by the Cirad cocoa program 

Montpellier. We are grateful to Touton Company, Indus-
trial private cocoa producer for providing cocoa pods and 

ADM cocoa for their logistical support and providing of 
uncontrolled cocoa samples. 
 

 
REFERENCES 
 
Awua PK (2002). Cocoa processing and chocolate manufacture in 

Ghana. Essex. David Jamieson and Associates Press Inc. 
Barel MA (1998). Première transformation du cacao. Formation de 

l’arôme cacao. In J. Pontillon. Cacao, chocolat, production, utilisation, 
caractéristiques. (pp. 96-115). Paris: Techniques et documentation, 
Lavoisier.  

EEC (1973). Directive 73/241/EEC by European Parliament and the 
European Council relating to cocoa and chocolate products intended 
for human consumption. Official Journal of the European Communi-
ties L 228 of 16/08/1973, pp. 0023-0035.  

Fowler MS (1999). Cocoa beans: from tree to factory. In: Beckett ST 
(Ed). Industrial chocolate manufacture and use, Oxford: Blackwell 
Science, pp. 8-35. 

Guénot MC, Perriot JJ, Vincent JC (1976). Evolution de la microflore et 
des acides gras des fèves de cacao au cours du stockage: étude 
préliminaire. Café, Cacao, 20: 53-58. 

Hagermann AE (1992). Tannin-protein interaction. In Ho CT, Lee CY, 
Huang MT (Eds). Symposium on Phenolic compounds in food and 
their effect on health: Analysis, occurrence and chemistry. ACS 
Series. 506: 237-247.  

Hiol A (1999). Contribution à l’étude de deux lipases extracellulaires 
issues de souches fongiques isolées à partir du fruit de palme PhD 
Dissertation, Université d’Aix, Marseille, France. 

Imeson HC, Rood SB, Weselake RJ, Zanewichn KP, Bullock BA, 

Stobbs KE Kocsis MG (1993). Hormonal control of lipase activity in 

oilseed rape germinant. Physiol. Plant. 89: 476-482. 

  
  

 
 

 
IOCCC. International Office of Cocoa, Chocolate and Sugar Confection. 

(1996). Determination of free fatty acids (FFA) content of cocoa fat as 
a measure of cocoa nib acidity. Analytical Method n° 42-1993. 

Kallithraka S, Bakker J, Cliffor MN (2001). Interaction of (+)-catechin, (-
)-epicatechin, procyanidin B2 and procyanidin C1 with pooled human 
saliva in vitro. J. Sci. Food Agric. 81: 261-268. 

Lipp M, Anklam E (1998). Review of cocoa butter and alternative fats for 
use in chocolate- Part A. Compositional data. Food Chem, 62: 73-79. 

Mert S, Dandik L, Aksoy HA (1995). Production of glycerides from 
glycerol and fatty acids by native lipase of Nigella sativa seed. Appl. 
Biochem. Biotechnol. 50: 333-342. 

Misnawi JS, Nazamid S, Jamilah B (2002). Activation of remaining key 
enzymes in dried under-fermented cocoa beans and its effect on 
aroma precursor formation. Food Chem. 78: 407-417. 

Nickless H (1996). Cocoa butter quality. In: Selamat J, Lian BC, Lai TK, 
Ishak WRW, Mansor M (eds). Proceeding of the Malaysian 
international cocoa conference Kuala, Lumpur. pp. 322-336. 

Ory RL, Saint-Ange AJ, Altschu AM (1962). The acid lipase of castor, 
properties and substrat specificity. J. Lipids Resour. 3: 99-105.  

Oyeniran JO (1980). The role of fungi in the deterioration of tropical 
stored products. Nigerian stored products research Institute (Lagos), 
Occasional paper series n° 2: 9-10. 

Oyeniran JO (1979). The influence of the storage environment on the 
quality of commercial cocoa with special reference to mouldiness of 

the beans. Proceedings of 7
th

 International Conference on cocoa 
research Douala, pp. 577-582.  

Poisson J, Cahagnier B (1979). Développement des moisissures sur les 
graines en fonction de diverses caractéristiques du substrat et de la 
population fongique. Tech. Indust. Céréalières, 169: 3-8. 

Pontillon J (1998). Le beurre de cacao et les matières grasses en 
chocolaterie. In : Pontillon J (ed) Cacao, chocolat, production, utilisa-
tion, caractéristiques (p. 326). Techniques et documentation, 
Lavoisier, Paris, pp. 257-269.  

Raghavendra MP, Prakash V (2002). Phenylcolboronic acid- a potent 
inhibitor of lipase from Oryzae sativa. J. Agric. Food Chem. 50: 6037-
6041. 
Renaud R (1954). La qualité du cacao. Les moisissures des fèves 

fermentées. Agronomie tropicale (Nogent-sur-Marne), 9 : 563-583. 
SAS Institute, Inc. (2002). JMP Software Version 5, Cary, NC.  
Selamat J, Hamid MA, Mohamed S, Man, CY (1996). Physical and 

chemical characteristics of Malaysian cocoa butter. In Selamat J, Lian 
BC, Lai TK, Ishak WRW, Mansor M (eds). Proceeding of the 
Malaysian international cocoa conference Kuala Lumpur. pp. 351-
357.  

Wanasundara PD, Wanasundara UN, Shahidi F (2001). Lipolytic activity 
of enzymes from germinating seeds of sesame (Sesamum incidcum 
L.). J. Food Lipids. 8: 75-84. 

Whitefield R (2005). Making chocolates in the factory. Kenedy’s 
Publications Ldt, London. 

 
Wood GA, Lass RA (1985). Cocoa. Logman Scientific and Technical, 

Essex.  
Yusep I, Jinap S, Jamilah B, Nazamid S (2002). Influence of 

carboxypeptidases on free amino acid, peptide and methylpyrazine 
contents of under-fermented cocoa beans. J. Sci. Food Agric. 82: 
1584-1592. 


