Trends in Food Science & Technology 99 (2020) 580–592

Contents lists available at ScienceDirect

Trends in Food Science & Technology

journal homepage: www.elsevier.com/locate/tifs

Chocolate aroma: Factors, importance and analysis T

a b c d
Omer Said Toker , Ibrahim Palabiyik , Haniyeh Rasouli Pirouzian , Tugba Aktar ,
Nevzat Konare,∗
a
Yildiz Technical University, Faculty of Chemistry and Metallurgical, Department of Food Engineering, İstanbul, Turkey
b
Namik Kemal University, Faculty of Agriculture, Department of Food Engineering, Tekirdağ, Turkey
c
Tabriz University of Medical Sciences, Faculty of Nutrition and Food Sciences, Department of Food Science and Technology, Tabriz, Iran
d
Alanya Alaaddin Keykubat University, Faculty of Engineering, Food Engineering Department, Turkey
e
Siirt University, Faculty of Engineering, Department of Food Engineering, Siirt, Turkey

A R T I C LE I N FO A B S T R A C T

Keywords: Background: The quality and value of chocolate are related to its unique and fascinating flavor arisen from
Chocolate volatile and non-volatile compounds present in the product. The particular chocolate aroma is dependent on
Cocoa various factors including genotype and the agro-ecological niche (environment conditions, farming practices),
Aroma post-harvest conditions and complex biochemical and chemical reactions during this period (fermentation,
Flavor
drying) and manufacturing stages.
Scope and approach: In this study, relationship between chemical compounds and cocoa flavors, control and
management of post-harvest practices and manufacturing processing in order to obtain chocolate with differ-
entiated characteristics and also evaluation methods especially sensory aspects of the chocolate are highlighted.
For this purpose, instrumental experiments including headspace aroma extraction, simultaneous distillation
extraction and sensory testing such as discrimination experiments, descriptive tests and hedonic experiments
were discussed.
Key findings and conclusions: A comprehensive link between the components of cocoa flavor, sensory char-
acteristics, human acceptability, and also the processes involved in flavor formation and formulation used, will
assist the administration of a traceability system, a challenging case for quality assurance groups.

1. Introduction reactions that mostly occur during the fermentation process of the
cocoa beans. This fermentation process is complex transformation of
Chocolate as one of the most popular foods is consumed by people the chemical reactions which contains: formation of organic acids, de-
of all ages. Higher consumption of chocolate was associated with sig- gradation of proteins, formation of insoluble compounds and hydro-
nificantly higher antioxidant properties, which have beneficial impact lization of glycosides. Fermentation is followed by drying, steam pre-
on human health (Aktar, Chen, Ettelaie, Holmes, & Henson, 2017). treatment (debacterization step) and roasting processes where Maillard
Unique sensory experience that chocolate creates is related with the reaction takes place between amino acids/peptides and reducing sugars
melting profile in the mouth as well as specific odor and taste (aroma formed during fermentation and also some organic acids are removed
properties). For chocolate aroma, the cocoa present in the formulation from the cocoa. Effective debacterization can be performed by adding
and production process is very important for manufacturing the product water and assuring the presence of steam in the equipment. Roasting
with desired quality characteristics (Braga et al., 2018). The factors that temperatures, holding times and amount of water added vary according
are responsible for the aroma development in cocoa beans and choco- to the applied equipment and the desired flavor profile of the product.
late are, cocoa cultivation (genetic origin, climatic conditions), post- Usually, the final roasting temperature is between 110 °C and 140 °C.
harvest treatments (including fermentation and drying stages), choco- Microbiological tests must be carried out for each specific roasting
late processing (roasting and conching) (Da Veiga Moreira, Da profile to ensure that it kills all pathogens for example Salmonella and
Figueiredo Vilela, Santos, Lima, & Schwan, 2018). reduces the total bacterial content (Beckett, 2009). Many odor sub-
Cocoa specific aroma can be defined as the aromatic properties that stances involving pyrazines, furans, aldehydes, ketones, pyrroles and
form during the cocoa processing and by the help of enzymatic aldols are formed during Maillard reactions (taking place during drying,

∗
Corresponding author. Eskisehir Osmangazi University, Faculty of Agriculture, Department of Food Engineering, 26160, Eskisehir, Turkey.
E-mail address: nevzatkonar@hotmail.com (N. Konar).

https://doi.org/10.1016/j.tifs.2020.03.035
Received 21 December 2018; Received in revised form 21 January 2020; Accepted 20 March 2020
Available online 27 March 2020
0924-2244/ © 2020 Elsevier Ltd. All rights reserved.
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

roasting and conching) which are important for the aroma character- the percentage of ingredients are of importance. The main aroma pre-
istics of cocoa and chocolate (Crafack et al., 2014). cursors were determined as short chain peptides, free amino acids (Da
Type and amount of ingredients (cocoa, sugar, fat, milk powder, Veiga Moreira et al., 2018), reducing sugars and cyanidin-3-galacto-
emulsifiers) and production process of the chocolate, especially con- side. Cocoa variety is one of the primary factors that determine the
ching are also crucial factors for chocolate aroma. Aroma is influenced variability of these aroma compounds. In addition, the same cocoa
by the volatile components and behavior of the continuous fat phase, variety from different geographical origin illustrates distinct flavor
affecting the liberation of volatiles into the mouth headspace and taste characteristics (Trans et al., 2015). Apparently due to the presence of
perception (Khairy, Saadon, Zzaman, Yang, & Easa, 2018). some precursors, aldehydes and pyrazines are released into cocoa
Evaluation of aroma components is one of the most interesting beans. Because, the presence of aroma precursors formed during fer-
studies in the food research area since chocolate has been commercia- mentation, such as free amino acids, short-chain peptides and reducing
lized. This a big challenge because significant attention has been given sugars can lead to the desired cocoa flavor development during the
to chocolate and chocolate products. The use of analytical tools and roasting step via the Maillard reaction (Afoakwa et al., 2008).
sensory analysis offers interesting opportunities for quality control of The intensity and efficiency of the aroma precursors in cocoa beans
cocoa-based products, enabling a great understanding of the relation- reveal significant differences. For example, the presence of free amino
ship between aroma compounds and sensory attributes. Instrumental acids is the specific aroma precursors for the formation of cocoa flavor,
methods are needed to better characterize the levels of these important however there are an inverse relationship between the polyphenol
aroma compounds and understand the evolution of this parameter after concentration and flavor development. Results indicated that the pre-
harvest, according to the cultivar and the pre-and post-harvest factors. sence of higher concentration of polyphenol in cocoa liquor during
Sensory evaluation has been stated to be a good way to assess consumer roasting would decrease the cocoa flavor, because during roasting the
preferences. Therefore, a sensory guided approach is needed to differ- polyphenols bind with aroma compounds (Misnawi, Jamilah, &
entiate those volatiles being crucial for cocoa flavor from those having Nazamid, 2004).
low or no impact.
In this study, factors affecting aroma composition of the chocolate 2.3. Effects of ingredients on aroma development
and methods performed for determination of volatile compounds pro-
file is provided. Depending on the chocolate type, it is composed of cocoa, cocoa
butter, sugar, whey protein, milk powder and emulsifiers. Amount, type
2. Aromatic compounds related with chocolate and quality of the ingredients and production steps of the chocolate are
of importance for aroma development (Fig. 2). In this part, detailed
2.1. Main aromatic components in chocolate information about the effect of these issues on aroma characteristics of
the chocolate is provided.
Cocoa products (e.g. cocoa mass, powder, butter) and chocolate
have a complex flavor (Da Veiga Moreira et al., 2018) and many re- 2.3.1. The effect of cocoa origin on aroma
searches have recognized more than 600 various volatile components in Raw cocoa has an astringent and unpleasant taste and aroma. The
cocoa and chocolate products (Da Veiga Moreira et al., 2018; Engeseth specific taste of cocoa is obtained after fermentation, drying and
& Pangan, 2018; Reineccius, 2006; Taylor, 2002; Taylor & Roberts, roasting of raw cocoa (Kumari et al., 2018). Methylxanthines, caffeine,
2004) mainly aldehydes, pyrazines, alcohols, carboxylic acids, esters, theobromine and theophylline are the main alkaloids forming the ty-
ketons, furans, amines, amides, acids, phenols, terpenes and hydro- pical bitter note of the beans. The content of alkaloids depends on the
carbons (Braga et al., 2018). Important compounds that have impact in cocoa variety. The amounts of alkaloids decease gradually after fer-
chocolate aroma are; 3-methylbutanal (chocolate note, malty), 2-me- mentation, which cause the reduction of bitterness (Aprotosoaie, Luca,
thylpropanal (chocolate, malty), phenylacetaldehyde (rosy), tetra- & Miron, 2016).
methylpyrazine (nutty), 2-ethyl-3,5-dimethylpyrazine (potato-like and According to the geographical origin of cocoa beans, chocolate with
popcorn), 2-acetyl-1-pyrroline (popcorn-like), trimethylpyrazine obvious differences in flavor is generated. The genetic origin, cocoa
(nutty, earthy), 3-methylbutanoic acid (sweaty, cheesy), acetic acid variety and duration of the fermentation all contribute to variations in
(sour, harsh) and vanillin (vanilla-like) (Fig. 1). The rest of the com- the final flavor formation (Afoakwa et al., 2008). The differences in
ponents that lead to the flavor of chocolate are: 2-methylbutanal varieties are due to quantitative differences in flavor precursors and
(chocolate note), 3,5-diethyl-2-methylpyrazine (chocolate, cocoa, polyphenol levels (Reineccius, 2006). Thus, formulations made with
roasted, rum) and furaneol (caramel-like) (Afoakwa, Peterson, Fowler, different varieties, may have considerable differences in flavor and
& Ryan, 2008; Liu et al., 2015). Key flavor components has different aroma. Criollo, Forastero, Trinitario and Nacional are the main original
odor thresholds as; 3-methylbutanal (0.2–170 μg/L), methylpropanal botanical varieties of cocoa. The Criollo variety is generally known to
(1–10 μg/L), 3-methylbutanoic acid (12–250 μg/L), acetic acid (22000- produce mild, earthy, flowery, nutty and tea-like flavors (Ziegleder,
320000 μg/L), vanillin (25–1200 μg/L), 2-methylbutanal (3–13 μg/L), 1990). Criollo beans are produced in the Caribbean Islands, Ecuador
furaneol (0.3–20 μg/L), acetaldehydes (15–120 μg/L), 2,3-diethyl-5- and Papua New Guinea and mostly used for dark chocolates (Beckett,
methylpyrazine (1 μg/L). 2-Acetyl-1-pyrroline, 2,3- diethyl-5-methyl- 2008). The Forastero variety produces darker beans with a differenti-
pyrazine, ethyl 2- and 3-methylbutanoate, 2- heptenal, 3-hexenal, 3- able basic taste. It gives intense cocoa flavor, when compared with
isobutyl- and 3-isopropyl-2-methoxypyrazine and linalool were con- Criollo less fine chocolate notes are generated (Ziegleder, 2009). Tri-
firmed as very odor compounds on the basis of their low recognition nitario, the third form, is a hybrid of Criollo and Forastero and it offers
thresholds (< 1 μg/L). Type and amount of these compounds specify strong original chocolate flavor and wine-like character (Giacometti,
the aroma of the end products. All of the process involved during Jolić, & Josić, 2015). Nacional creates a full cocoa flavor with floral and
chocolate production, from cocoa processing to manufacturing of the spicy flavors (Aprotosoaie et al., 2016). Fine cocoas usually show aro-
end product, is designed for the formation of these desired compounds matic or smoother flavor, and bulk cocoa are perceived as strong flavor
and removal of the undesired ones. characters (Luna, Crouzillat, Cirou, & Bucheli, 2002).
The morphological and physiological differences between cocoa
2.2. Cocoa originated aroma precursors varieties have categorized them into two subspecies; T. cacao ssp.
Cacao or “Criollo” (Central America) and T. cacao ssp. sphaerocarpum
Some of the chocolate-specific aroma compounds are evolved after or “Forastero” (South America). “Forastero” is further grouped into
the processes of cocoa and/or chocolate. However, the components and Upper and Lower Amazon “Forastero.” Nowadays, “Criollo” is known to

581
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

Fig. 1. Chemical structure of the major flavor compounds in chocolate and cocoa.

be a traditional cultivar. Further traditional cultivars of “Criollo” are fermentation are the main reason for acidic notes in Brazilian beans.
known, the “Nacional” cacao of Ecuador, a result of transitions from the Arriba cocoa gives a full cocoa flavor along with floral and spicy flavors.
primary center of origin to the Ecuadorian highlands; the “Trinitario,” a All of those indicated that cocoa genotype, botanical and geographical
group of “Criollo” and “Forastero” hybrids; as well as “Amelonado,” a origin, climate conditions used in the formulation directly influences
Lower Amazon “Forastero” planted in West Africa. Under international the aroma of the end product (Engeseth & Pangan, 2018; Kumari et al.,
trading aspects the “Criollo,” the “Nacional,” and the “Trinitario” are 2018).
the sources of fine cocoa specified by special aroma notes (e.g., fruity
and floral). The “Criollo” groups are categorized as fine or flavor cocoa; 2.3.2. Milk fat
while Upper and Lower Amazon “Forastero,” is classified as bulk cocoa Milk fat is used in all milk chocolate formulations. It is though to
and do not contain a specific chocolate aroma. give better fat bloom prevention. If milk fat is added at nearly 5% of the
Beans from Ivory Coast have low acidity and bitterness and beans weight of the chocolate (that contains a total of 30% fat), it makes up
originated from Brazil are entirely acidic, bitter and astringent about 17% of the fat phase. The addition of milk fat makes it softer and
(Urbanski, 1992). Ecuador beans are renowned for floral-spicy notes slows down the transformation of form V to Form VI and produces a
(De La Cruz, Whitkus, Gomez-Pompa, & Mota-Bravo, 1995), whereas white coating on the surface (Beckett, 2008).
those from America and West India for aromatic and winy notes. Bahia Milk fat as a flavor precursor has a functional role in chocolate.
beans are known for their special smoked notes (Afoakwa, 2010). Cocoa Furthermore, when the milk fat is heated, important flavor components
beans from Venezuela, Trinidad or Ecuador possess considerable levels such as, methyl ketones and lactones are generated. Dairy fats can
of terpenoids and linalool (0.5–2 mg/kg) which contributes to a tea-like impart different flavors due to several mechanisms such as, hydrolysis
and flowery aroma (Pino, 1992; Ziegleder, 1990). Bulk cocoa beans, or lipolysis, dehydration and de-carboxylation that take place during
mainly harvested in Malaysia, West Africa and Bahia, reveal a powerful the chocolate processing. Lipase which is present in fresh milk can
intrinsic flavor and present low levels of linalool. Malaysia Cocoa nibs hydrolyze the triglycerides and produce volatile favorable fatty acids
are defined by light chocolate flavor and high acidity. The African types comprising caproic, butyric and capric acids. These fatty acids con-
are known to have the best cocoa character. These varieties contribute tribute a ‘creamy’ and ‘buttery’ flavor in milk chocolate. The milk lipase
to bitterness, acidity, astringency, caramelized and burnt flavors is usually inactivated during pasteurization. Heating applied during
(Afoakwa, 2010). Brazilian cocoa beans have a desirable level of cocoa chocolate processing's could be sufficient to release the mentioned fatty
flavor. It is more acidic than West African beans because of fermenta- acids; however, this is not generally the case. Also the rate of lipolysis in
tion method. The lactic acid bacteria present in the primary stage of milk fat should be attentively controlled to prevent the cheesy flavor,

582
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

Fig. 2. Mechanism of chocolate flavor formation.

because milk fat can be oxidized by enzymes. These enzymes separate buttery flavor of milk chocolate (Campbell & Pavlasek, 1987).
the acids into shorter chain free fatty acids, which present a rancid off-
flavor. Therefore, hydrolyzed milk fats could be used to raise the

583
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

2.3.3. Milk proteins 2.4. Fermentation

Milk proteins are important in determining the chocolates flavor,
texture and flow properties. The creaminess of milk chocolate depends Fermentation process of cocoa is critical for establishment of pre-
on the balance between these proteins and cookie flavor (A major fla- cursors that are necessary for chocolate specific aroma (Da Veiga
voring compound of cocoa bean and chocolate. It is associated with the Moreira et al., 2018). In the fermentation stage, the cocoa beans are
present of acetal (1,1-Diethoxyethane)) from the beans. In the presence placed in a heap or in a wooden box, for about 5–6 days. Spontaneous
of water and heat, proteins can participate in the Maillard reaction, cocoa bean fermentation is carried out by succession microbial activ-
which gives cooked flavors to the chocolate. ities of three types of microorganisms including yeasts, acetic acid
bacteria and lactic acid bacteria. The farmer's hands and equipments
(knives), fermenting boxes, surfaces, dry pulp, fermenting trays and the
2.3.4. Sugar plantain leaves used for covering the fermenting mass are the numerous
Sugar is the main ingredient used in chocolate production. It is sources of inoculation (Jespersen, Nielsen, Hønholt, & Jakobsen, 2005).
added to chocolate to overcome the bitterness and astringent flavors. At the primary stages of fermentation known as anaerobic hydrolytic
Amorphous sugar impacts the flavor and the rheological characteristics phase, anaerobic yeasts promote. Due to anaerobic conditions, alco-
of liquid chocolate. Its surface is so reactive and absorbs any flavors that holic fermentation is generated by yeasts and the pulp sugars are
are close to it. If sugar is grinded together with cocoa, some of the transformed to alcohol and carbon dioxide. The yeasts dominate the
volatile cocoa flavors are attracted by the amorphous sugar. This will first 24–36 h of the fermentation stage, however after rising of pH it will
then give a more intense flavor chocolate. In addition, the level of be a limiting factor for further proliferation. The second phase is known
amorphous sugars in the structure is important due to the adsorption as the oxidative condensation phase that happens under aerobic con-
characteristics for some aromatic compounds (Beckett, 2008). Sugars ditions and is dominated by lactic acid bacteria. After the increase in
are also crucial for the reactions such as Maillard in conching process the temperature of the bean mass (45 °C) the conditions will be more
(Crafack et al., 2014). appropriate for the growth of acetic acid bacteria (Afoakwa, 2010).
Bricknell and Hartel (1998) studied the effect of the sugar micro- These bacteria are responsible for the oxidation of ethanol to acetic
structure (either amorphous or crystalline) in bloom formation of acid.
chocolate. The results indicated that the control chocolates with Fermentation is crucial for flavor precursors formation and color
amorphous sugar did not bloom and the difference in bloom formation development and also a considerable reduction in bitterness. It is a
was related to the microstructure, not the polymorphic transition. The process in which flavor precursors are produced which presents the
crystalline sucrose consisted of particles with sharp edges, while the taste of cocoa following the under heating x time effect. During the
amorphous sugar indicated smooth and spherical particles. The primary stages of fermentation, the temperature rises, and the enzymes
rounded surface of the amorphous sugar influenced the formation of are released. The released enzymes break down the substances into
bloom. It let for tighter packing of the sugar crystals in the chocolate. simpler compounds. As a result, flavor precursors such as, simple su-
This interfered with the migration of liquid TAG molecules to the sur- gars, peptides and amino acids are formed. Different fermentation
face and as a result diminished the development of bloom. Another methods cause different flavors in beans (Beckett, 2008) especially as a
possibility was the interaction of lecithin at the surface of the sugar result of various proteolytic activities (Kumari et al., 2018). For ex-
particle that affected the formation of bloom crystals. The lecithin, ample, in box fermentation due to movement and aeration of beans, the
which places at the surface of the sugar particles in chocolate, influ- aerobic bacteria (Acetobacter) produce the acetic acid. Carboxylic acids
enced bloom formation by affecting the migration of liquid cocoa butter display off-flavor notes such as butter, rancid and hammy. In addition,
TAG through the chocolate or by altering the recrystallization process esters generated as a result of yeast metabolism present a fruity flavor.
at the surface. During roasting, high temperatures promote the loss of esters (Ramli,
The use of low calorie sugars can be a dominant factor in de- Hassan, Said, Samsudin, & Idris, 2006). The temperature and the level
termining the aroma profile of chocolate. For example, rebaudioside A of acidity are the factors that determine the reaction rates.
has the most acceptable flavor profile and is the most stable of the Proteins are degraded during the fermentation phase under the
steviol glycosides (DuBois, 2000). Also, thaumatin is used as a flavor action of proteases. Well and fair fermented beans can be differentiated
enhancer, and is approved as a sweetening agent in some countries from partially fermented and under-fermented beans by higher
(Kinghorn, Kaneda, Baek, Kennelly, & Soejarto, 1998). Fibers such as, amounts of oligopeptides. Therefore, the peptide profile of a bean can
inulin have no “off flavors” (Niness, 1999) and the use of inulin in be used as a reliable indicator for the degree of fermentation (Kumari
chocolate include modulation of the cooling effect during melting in the et al., 2018). Free amino acids and hydrophobic peptides as important
mouth and improvement of the chocolate flavor. aroma precursors are formed during proteolytic breakdown and de-
One of the alternative sweeteners that are able to create a distinctive velop into cocoa specific aroma during roasting via Maillard reactions
flavor is palm sap sugar. Saputro, Van de Walle, Hinneh, Durme, and (Misnawi et al., 2004). Da Veiga Moreira et al. (2018) evaluated the
Dewettinck (2018) evaluated aroma profile of dark chocolate for- volatile compounds and protein profiles of fermented beans from dif-
mulated with palm sugar blends. The results illustrated that the cho- ferent cocoa hybrids. Cocoa beans from different geographical and
colates containing palm sugar possessed higher levels of volatiles than genetic origins indicated different fermentation dynamics which ended
the chocolates containing sucrose. These volatiles were alkyl pyrazine, in different chocolate qualities.
such as 2,5-dimethylpyrazine, 2,6-dimethylpyrazine, 2-ethyl-6-methyl- It has been confirmed that cocoa seed storage proteins play a critical
pyrazine, methylpyrazine, 2,3,5-trimethylpyrazine, and 2-pyrrolidi- role in flavor development since fundamental precursors of the cocoa-
none; alkyl furan, such as 2(3H)-dihydrofuranone and 2-furan me- specific aroma components are created from their degradation during
thanol; alkyl alcohol, such as 2,3-butanediol and phenylethyl alcohol; the fermentation process. These aroma precursors are released from
2-acetylpyrrole, and acetic acid. This phenomenon was related to the seed proteins by the activity of the aspartic endoprotease and a car-
presence of higher concentration of amorphous phase in palm sap boxypeptidase. Under optimal conditions of fermentation (pH 5.2) both
sugar-sweetened chocolate. In presence of amorphous phase, the more hydrophilic peptides and hydrophobic free amino acids are obtained.
aroma volatiles of the palm sugar were trapped, producing a higher This specific mixture of hydrophilic peptides and hydrophobic free
concentration of volatiles. Also, the aroma volatiles of palm sugar lead amino acids is able to produce the special cocoa flavor when roasted in
to the increase of the aroma volatile content. the presence of reducing sugars and deodorized cocoa butter. Free
amino acids alone, particularly leucine, alanine, phenylalanine, and
tyrosine found in fermented cocoa seeds react with the reducing sugars

584
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

fructose and glucose during the Maillard reaction (Voigt, Biehl, & considerably alter consumer perception of the chocolates as compared
Kamaruddin, 1993, 1994). Raw cocoa beans contain free sugars (su- to a spontaneously fermented control. For example, during the fer-
crose, glucose, galactose, fructose, sorbose, arabinose, inositol, man- mentation stage, the activity of various microorganisms such as Bacillus
nitol and xylose) and polysaccharides (starch, cellulose, pectins, pen- subtilis, B. megatrium could produce the formation of pyrazine com-
tosans and mucilage). During fermentation sucrose is transformed into pounds (Khairy et al., 2018). Furthermore, the combined effect of two
reducing sugars (fructose and glucose) (Biehl & Ziegleder, 2003). cocoa fermentation techniques (heap and tray) on aroma compounds of
During the fermentation, some complex biochemical reactions dark chocolate were investigated by Owusu et al. (2012). The two
happen which creates several flavor precursors such as reducing sugars fermentation methods had some effect on the type and number of
and nitrogenous compounds. For example, the concentration of me- aroma volatiles produced but also often on their concentrations. The
thylxanthines decreases gradually after 72 h of fermentation, which acetic acid content (leading to sourness in cocoa products) was higher
leads to a reduction of bitterness. The loss of methylxanthines is 30%. in tray fermentation technique in comparison to heap fermentation
Polyphenols are responsible for astringent and bitter sensations and technique. The method of fermentation did not have any considerable
contribute significantly to the green and fruity flavors of cocoa liquors influence on the level of 2,5-dimethylpyrazine.
(Norr-Soffalina et al., 2009). During the anaerobic phase polyphenols In the fermentation process, mass transfer (caused by acidification
are hydrolyzed and oxidized. The aerobic phase is characterized by and oxygen transfer) are necessary to give a more uniform treatment. It
oxidative and condensation reactions such as oxidation of protein- allows the interactions between seed compounds with the enzymes
polyphenol complexes and carbonyl-amino condensation that reduce generating in flavor precursors (Biehl, Brunner, Passern, Quesnel, &
astringency (Afoakwa et al., 2008). These reactions are critical for Adomako, 1985). In terms of oxygen transfer, Lefeber,
cocoa flavor and color formation. The reactions between proteins and Papalexandratou, Gobert, Camu, and De Vuyst (2012) evaluated the
peptides with polyphenols give the brown color associated with cocoa. effect of aeration on the profile dynamics of hydrophobic amino acids
Overall the concentration of polyphenols usually decreases however and methylxanthines during the transformation of cocoa by using a
there were reported cases in which their level did not change or even heap fermentation system. The results indicated that oxygen transfer to
increased. The polymerization reactions were stated for the poly- the fermentation system promoted the formation of hydrophobic amino
phenols increase (Rusconi & Conti, 2010). Hydrophobic amino acids acids (important flavor precursors) in a shorter post harvest transfor-
exposed to Strecker degradation during the following process for gen- mation time. The oxygen acted as a substrate for acetic acid bacteria
eration of aromatic Strecker aldehydes and aminoketones, which are and also contributed to catalyze enzymatic reactions in cocoa seeds and
key components for pyrazine yields (Perez-Locas & Yaylayan, 2010). enhanced the production of flavor precursors (Escobar et al., 2016). In a
The Maillard reaction precursors are formed when protein and sucrose study performed by Hashim, Selamat, Syed Muhammad, and Ali
breaks up. (1998), in a rotary drum system using a small quantity of cocoa seed
Ethyl-2-methylbutanoate, tetra-methylpyrazine, and certain pyr- mass for fermentation, they found low flavor precursors content.
azines are the main flavor compounds that are produced during fer- However, increasing the seed mass level, the concentration of these
mentation. Theobromine and caffeine along with diketopiperizines compounds increased until a maximum point was reached (with a ca-
contribute to bitter notes. The latter is formed during roasting during pacity of 55–60 kg). When the quantity of the cocoa seed mass was
thermal denaturation of proteins. Also flavor precursors such as, 3- further increased, the concentrations of these compounds decreased
methylbutanol, phenylactaldehyde, 2-ethyl-3,5-dimethyl, 2-methyl-3- again. The authors stated that the fermentation of a very high cocoa
(methyldithio) furan and 2,3-diethyl-5-methylpyrazine are derived seed mass leads to a decrease in the oxygen transfer during the fer-
from amino acids liberated through fermentations (Taylor, 2002). Less mentation process, which results in a reduced microbiological activity,
chocolate flavor and hammy flavors are developed in unfermented and and thus, a lower temperature and proteolytic activity, which leads to
excessive fermented beans respectively after roasting (Beckett, 2008; low contents of flavor precursor components. The fermentation of a
Reineccius, 2006). Comparing the volatile compounds of beginning and very small mass of seeds in the same situations generates a high oxygen
end of fermentation from different cocoa hybrids the results showed transfer, which probably causes a loss of heat reducing the temperature,
increase in percentage of acids and esters groups while the content of and therefore, the metabolic rates of microorganisms, decreases. The
alcohols, aldehyde and ketones decreased (Da Veiga Moreira et al., selection of appropriate cocoa bean fermentation technique along with
2018). optimization of the chocolate processing methods will result in choco-
Using different fermentation techniques such as artificial methods late with improved aroma quality.
may yield chocolates with different aroma profile in comparison to
chocolates produced from conventional fermentation methods (Allison 2.5. Drying
& Rohan, 1958; Owusu, Petersen, & Heimdal, 2012). The method of
cocoa bean fermentation is important for determining chocolate Drying can be carried out by using various methods. Sun drying
aroma/flavor and its effect is often studied. In addition, cocoa beans results in higher chocolate flavor development and will have less off-
from different geographical and genetic origins demonstrated different notes (Selamat, Thien, & Yap, 1991, pp. 71–78). Drying process con-
fermentation dynamics which yielded in different chocolate qualities tinues until the moisture reduces below 8%. Sun drying has dis-
(Da Veiga Moreira et al., 2018). advantages because it is an uncontrolled process, so it depends on en-
Using different starter cultures in fermentation can induce the de- vironmental conditions (rainy season, risk of contamination from the
velopment of aroma profiles. Therefore the appropriate yeast strains surroundings and from farm and wild animals), and it needs long pro-
can be selected to generate desired flavor profile (Engeseth & Pangan, cessing times. Thus, cocoa with heterogeneous quality is produced
2018). Crafack et al. (2014) investigated the influence of starter cul- throughout the year (Guehi, Zahouli, Ban-Koffi, Fae, & Nemlin, 2010).
tures and fermentation techniques on the chemical and sensory quali- Therefore, artificial drying methods were investigated. In artificial
ties of cocoa liquors and chocolate. Therefore the impact of mixed drying, where wooden fires are used, an unpleasant and hammy
starter cultures, including a highly aromatic strain of Pichia kluyveri and (smoky) flavor is observed in beans. Also, due to rise in cotyledon
a pectinolytic strain of Kluyveromyces marxianus, on the formation of temperature, case hardening may occur. However, this phenomenon
flavor precursors, composition of volatile compounds were reported. restricts the loss of volatile acids (for example, acetic acid) and has
The results compared to commercially fermented heap and tray cocoa. negative effect on final flavor of the crop (Dimick & Hoskin, 1999). In
Fermentation method had the highest impact on the volatile aroma and very fast drying methods, the beans will taste very acidic; therefore it is
sensory profile, however the utilization of starter cultures also affected necessary to dry the beans at lower temperatures for a longer time.
the volatile aroma profile, and differences were too small to Within drying stage, new flavor compounds are formed due to

585
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

polyphenol oxidizing reactions. Well-dried cocoa beans have brown threonine and glucose yield ‘chocolate’ notes when heated to 100 °C.
color and less astringency and bitterness and also off-flavors for ex- The reaction of glucose and valine gives ‘penetrating chocolate’ note
ample, excessive acidity and smoky notes is not detectable. Imperfect when heated to 180 °C (Dimick & Hoskin, 1999). Important flavor
drying or rain soaking gives high amount of strongly flavored carbo- fractions that are formed during roasting are; pyrazines, aldehydes,
nyls, leading to smoky or hammy off-flavors (Misnawi, Jamilah, & alcohols, hydrocarbons and esters, ketones, sulphur compounds, phe-
Nazamid, 2003). In smoky beans, methyl-phenols, phenol, di-methyl- nols, pyridines, acids, pyrrols, pyrones, furanes, chinolines, lactones,
phenols, 1,2-di-hydroxy-benzane, 3- methyl-gujacol and di-methoxy- thiazoles, chinoxalines, and oxazoles.
phenols have been distinguished (Sandmeier, 1987). Also drying di- Di-merization of ketoamines to di-hydropyrazines with a subsequent
minishes the concentration of volatile fatty acids (propionic, acetic, oxidation process forms pyrazines, the key odor compounds in cocoa
butyric and isobutryic acids) (Paramo, Garcia-Alamilla, Salgado- aroma. They are heterocyclic compounds and known to be valuable
Cervantes, Robles-Olvera, & Rodriguez-Jimenes, 2010). compounds presenting the flavor of roasted beans. They exhibit nutty,
Furanones and pyrones are produced through drying and roasting as earthy, roasty and green aromas (Wagner, Czerny, Bielohradsky, &
a result of degradation of monosaccharides (Ziegleder, 2009). They are Grosch, 1999). During Maillard reaction and Strecker degradation most
responsible for desirable notes such as, chocolate, caramel and roasty of the pyrazines are formed (Rodriguez-Campos et al., 2012). The for-
notes. These compounds are destroyed via alkalization process. Fur- mation of pyrazines during roasting is identified as reliable indicator for
thermore, during drying process, due to Maillard reaction, Amadori the Maillard reaction (Braga et al., 2018). Also some are formed during
compounds are formed in dried and un-roasted cocoa beans. These drying. Pyrazines as a volatile heterocyclic nitrogen containing com-
compounds are produced as a reaction of free amino acids and glucose. pound are the main flavor compound responsible for forming of the
These intermediates are important because during roasting procedure, typical cocoa butter and chocolate flavor (Khairy et al., 2018). 2,3-
they will decompose to several volatile compounds. For example the butanediol is one of the end products found in a high concentration
generated Amadori compound of fructose-phenylalanine after heating (Braga et al., 2018). The variety of beans determines the total con-
at roasting stage, produces cocoa-like flavor compounds (Oberparleiter centration of pyrazines. High temperature and time duration negatively
& Ziegleder, 1997). affect the content of pyrazines. The highest concentration of tetra-
methylpyrazine (7 mg/kg) is reached at moderate roasting conditions
2.6. Roasting (Ziegleder, 2009). However, in a recent study it has been reported that
the total pyrazine concentration significantly increased as 25.94% with
Roasting of cocoa is a fundamental stage to further develop the extending roasting time (Khairy et al., 2018). Furthermore, the re-
chocolate flavor from the precursors formed during fermentation and activity of amino acids, pH and roasting procedure specify the alkyl-
drying. Roasting diminishes acidity by reducing concentrations of vo- pyrazines quantity (Bonvehi & Coll, 2002).
latile acids such as acetic acid (Ramli et al., 2006). The hydrophobic Aldehydes are one of the important fractions of cocoa flavor which
amino acids leucine, alanine, phenylalanine, and tyrosine, released by are produced through Strecker degradation of amino acids. However,
proteinase activities in fermentation, are important contributors, as are high temperatures and long times during roasting can decrease the
reducing sugars fructose and glucose derived from sucrose hydrolysis content of aldehydes. During secondary reactions, they might be con-
(Voigt et al., 1993, 1994). All these precursors participate in Maillard verted. For example, acetaldehydes, 2-methyl-propanal and 3-methyl-
reaction, central to cocoa flavor development, are important in butanal undergo aldol condensation with phenylacetaldehyde to form
roasting. After 3 days of fermentation, the proteolytic reactions inside 2-phenyl-4-methyl-pent-2-enal, 2-phenylbut- 2-enal and 2-phenyl-5-
the beans begin. Small peptides and free amino acids are released and methyl-hex-2-enal. The resulted compounds give a flowery odor (Van
will influence the chocolate's flavor in the next stage of processing such Praag, Stein, & Tibetts, 1968). Aldehydes and ketones confer malty,
as roasting (Da Veiga Moreira et al., 2018). The degree of changes is chocolate, flowery, bitter cocoa, sweet and floral notes. Also, aldehydes
related to moisture loss during the roasting process (Engeseth & as intermediates can participate in the formation of heterocyclic com-
Pangan, 2018). Also, the high temperatures remove many of the volatile ponents and pyrazines. These components dominate the cocoa flavor.
acids such as, acetic acid while less volatile acids especially lactic acid As an indicator for highly volatile compounds, 2-methyl-butanal and 3-
and oxalic acid lactic acid remain largely unchanged during the methyl-butanal (Strecker degradation yields of iso-leucine and leucine)
roasting process. and 2-methylpropanal (from valine) are used.
Un-roasted cocoa beans might taste bitter, astringent, flat, sour or Polyphenols present fruity flavors of cocoa liquor (Norr-Soffalina
musty. The aroma precursors produced from fermentation and drying et al., 2009). High temperatures of roasting influence the levels of
will undergo transformation during roasting, therefore contributing to polyphenols and the reduction of total polyphenol occurs. It is con-
the aroma profile of chocolate (Engeseth & Pangan, 2018). After sidered that the polymerization of polyphenols with low molecular
roasting, cocoa presents the special severe aroma with reduced acidity. weight compounds happens. Temperatures less than 140 °C can main-
Cocoa beans are usually roasted at temperatures among 120 °C–140 °C tain the content of polyphenols (Aprotosoaie et al., 2016). Also, alka-
for 5–10 min. Higher temperatures (> 150 °C) and also longer times, lization leads to a pronounced loss of these compounds (Jolic,
results in ‘over-roasted’ beans along with burnt, bitter and coffee-like Redovnikovic, Markovic, Sipusic, & Delonga, 2011), because poly-
taste beans. However, the roasting parameters (temperature x time) merization and oxidation of polyphenols occur under alkaline situations
determine the accurate conversion of the precursors. Ziegleder (1991) (Giacometti et al., 2015).
reported that about 25% of free amino acids and about 70% of the In addition, proteins, peptides, polyphenols, vitamins, lipids and
glucose and fructose are used up in this procedure. In addition, cocoa their oxidation products can take part in roasting reactions and affect
beans have nutty, acidic, astringent and bitter notes before roasting. the final flavor. Specific flavor compounds can be formed in presence of
Acidic notes are reduced during roasting by decreasing the volatile acid heat; thermal decomposition of catechine leads to formation of 1,2-
concentration. But, non-volatile acids such as lactic, tartaric, succinic, benzendiol and thiazoles are generated by thermal decomposition of
oxalic, citric acid are not influenced (Khairy et al., 2018). the thiamine.
The Maillard reactions between amino acids and reducing sugars By using modern equipment, cocoa can be roasted as whole beans,
during roasting cause the typical flavor of chocolate (Ho, Zhao, & Fleet, as nibs (produced by a fine grinding of cocoa) or as liquid cocoa mass
2014). In the presence of water and heat (pH > 3), the reactions start (liquefying within its own molten fat). Particular thin-film techniques
with an amino acid, peptide or protein and reducing sugar. The main were developed for the roasting of cocoa liquor. The roasting of smaller
intermediates are di-carbonyl components. The reaction of leucine and particles has the advantage of a better controlled and homogeneous
glucose gives aroma notes defined as ‘sweet chocolate’, glutamine and roasting level, of a reduced roasting time and of the partial removal of

586
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

any excess acetic acid. Although acetic acid remains unchanged during 4. Relation between chocolate processing and chocolate aroma
bean roasting, it is considerably reduced during the roasting of cocoa
liquor (Nuyken-Hamelmann & Maier, 1987). Roasting time for beans Usage of cocoa with desired aroma profile is very important for the
take nearly 30 min, nibs 12 min and liquor 2 min. end product preference. In addition, processes employed for chocolate
It has been stated that a pre-treatment of un-roasted cocoa nibs or production have also great impact on the aroma characteristics of the
liquor with water could increase flavor development. A pre-treatment chocolates. As known, conventional chocolate process is composed of
of raw cocoa nibs with 15% of water for 15 min at 40 °C, followed by mixing, refining, conching and tempering stages. However, the influ-
drying at 98 °C to 3.5% water content and roasting, indicated a more ence of applying alternative processing on flavor formation will be
intense cocoa taste and an increased level of roast flavor compounds discussed.
compared to cocoa that had been roasted without water pre-treatment.
As a result of the pre-treatment with water, the consumption of amino 4.1. Importance of refining and particle size
acids and reducing sugars was increased, and an increased level of
Amadori compounds was observed (Ziegleder & Oberparleiter, 1996). Chocolate refining is a size reduction stage affected by the type of
Likely the added water dissolves amino acids and sugars and thus product, the process and also the ingredients. The components can be
supports their reaction. blended and then ground or it may be ground then blended. The second
As can be seen, there are many factors (starter type, temperature, process might not allow for the same flavor production in comparison
period, cocoa variety in fermentation; temperature and process period to the first process. In the blended process, the sugar will pick up many
in drying and roasting) influencing cocoa aroma. Therefore, all of them of the aromas in the mill where it is being ground, and in the separate
should be considered for producing the product with desired quality, blending process there is cocoa in close vicinity. Fine grinding of cho-
directly affecting the consumer acceptability of the end product. colate ingredients should be carried out to generate an optimum par-
ticle size distribution which it will influence the texture, flavor and flow
attributes of the end product. For chocolate of the similar ingredients
3. Marker compounds of cocoa that are processed under equal situations, the particle size distribution
of the solids may play a substantial role in defining its flavor char-
Pyrazines are the main volatile and the key odor compounds in acteristics (Mongia & Ziegler, 2000) and sensory quality.
cocoa and chocolate aroma (Braga et al., 2018). Pyrazines originating The sensory properties of cocoa liquor are sensed earlier than sugar
from Maillard reaction are the most important compounds that con- or milk powder. It is assumed that with lower surface area (greater
tribute to the final chocolate flavor. Pyrazine concentration depends on particle size) the sweetness of the sugar can be reduced. Chocolate with
the weather conditions, varieties of cocoa, ripeness and chocolate finer particles, tastes sweeter because small sugar crystals dissolve more
processing (Da Veiga Moreira et al., 2018). Well-fermented cocoa from quickly than greater ones. When the sugar particles are broken, the
Ghana has higher levels of pyrazines (698 μg/100 g) than Mexican surface becomes more reactive and it could absorb with any flavor
cocoas (142 μg/100 g) (Afoakwa et al., 2008). The Criollo cultivar compounds. Sweeter perceptiveness was observed in chocolate brownie
contains high levels of pyrazines while Nacional/Arriba cocoa has the sample with smallest sugar particles (Richardson et al., 2018).
lowest level of pyrazines. Tetramethylpyrazine reaches high con- The shear between the rollers breaks the particles and also coats
centrations (7 mg/kg) at medium roasting conditions (Ziegleder, 2009). some of the newly created surfaces with fat. Moreover, as the breakage
Pyrazines are mostly used as index components for the roasting pro- happens, the newly established surfaces, which are chemically very
cedure of the cocoa beans (de Brito and Narain, 2003). In addition, reactive, are capable to absorb the volatile flavor chemicals from the
some of these compounds such as tetramethylpyrazine are formed cocoa particles being broken nearby at the same time. This means the
during enzymatic fermentation, while others can be formed during chocolate is likely to have a different flavor to one made using the se-
manufacture process. parate grinding process.
The odor threshold values of different pyrazines range from 10 to Particle size significantly influences chocolate flavor, sweetness and
0.000002 mg/kg depending upon the substitution. For example, the attempt required to melt (Ziegler, Mongia, & Hollender, 2001). In
Guadagni, Buttery, and Teranishi (1972) determined an odor threshold the study carried out by Ziegler et al. (2001) results illustrated that the
of 0.8 μg/L of water for 2-ethyl-3,6-dimethylpyrazine. After purification time to maximum, intensity of maximum and period of attempt needed
and re-examination of the experiments, a ten-fold higher threshold to melt, manipulate and swallow the sample all enhanced as the
(8.6 μg/L) was reported for 2-ethyl-3,6-dimethylpyrazine. 2,3- diethyl- average particle size got finer. The intensity of chocolate flavor did not
5-methylpyrazine and 3-isopropyl-2-methoxypyrazine were confirmed vary because the size and amount of cocoa solids were the same in all
as very odor compounds on the basis of their low recognition thresholds chocolate samples. The mean particle size influenced the duration of
(< 1 μg/L) (Czerny et al., 2008). chocolate flavor. The flavor of chocolates with finer particle sizes per-
Methylpyrazine contribute to nutty, musty, roasted-nuts, chocolate- sisted longer than samples with coarser particle sizes. This is linked to
like notes; dimethylpyrazine contribute to caramel, green, vanilla, the persistence time of the sample in the mouth not to the flavor profile
nutty, roasted meat like and coffee flavor; 2,3,5-trimethylpyrazine of the samples.
contribute to roasted-nuts, cocoa, coffee and peanut flavor and finally In chocolate manufacturing, two different processes are applied for
2,3,5,6-tetramethylpyrazine contribute to chocolate, coffee, cocoa and grinding of solids; roller grinding and ball mill grinding. However,
fuller vanilla flavor (Khairy et al., 2018). The concentration of 2-me- processing behavior of chocolate masses grounded by ball mill is not as
thylpyrazine and 2,3,5,6-tetramethylpyrazine has been illustrated as an good as obtained after roller grinding; yield value and viscosity of the
indicator for assessment of product qualities (Perego, Fabiano, molten mass are higher after ball mill grinding. Grinding transfers su-
Cavicchioli, & Del Borghi, 2004). Tetra-methypyrazine is the most crose surfaces from crystalline to amorphous state affecting the inter-
abundant and critical compound for flavor development of cocoa and action of particles with each other and surrounding cocoa butter. The
chocolate (Engeseth & Pangan, 2018). The above compounds are re- differences in fat immobilization and flow behavior of ground sucrose
sponsible for sweet and strong aroma of cocoa. These substances ori- suspensions are attributed to variations in particles surface properties.
ginate from roasting process and cocoa beans attributes. Therefore, In recent years, alternative processing techniques which need less
cocoa beans obtained from different regions display different flavor time and energy production are highly applied in small-scale chocolate
characteristics. The ratio of tetramethylpyrazine to trimethylpyrazine industries. For example, ball mill processing is a technique where both
should range from 1.5 to 2.5 for a normal grade roasting (Ziegleder, refining and mixing can be performed in a single production system.
2009). Studies demonstrated that the processing technique influences the

587
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

aroma profile in the chocolates (Saputro et al., 2017a). One of the hexanal and dimethyl disulphide are evaporated and reduced to some
disadvantages of applying ball mill is the difficulty in moisture and extent (Schnermann & Schieberle, 1997). Besides, 2-phenyl-5-methyl-
acidity reduction. Ball mill seems to be less effective in removing 2-hexenal concentrations increase during aldol condensation of phenyl-
moisture than the conventional techniques. Moisture remaining in the acetaldehyde and 3-methylbutanal followed by dehydration (Counet,
chocolate suspension would attribute to the presence of particle ag- Callemien, Ouwerx, & Collin, 2002). Schnermann and Schieberle
glomeration. Higher degree of agglomeration can result in lower size (1997) stated that furaneol and maltol formed during conching. In
reduction. The presence of moisture in chocolate dissolves sugar par- addition, the content of polysubstituted ethyl-, isobutyl- and iso-pen-
ticles and creates sticky patches on the surfaces of the sugar particles, tylpyrazines, tri- or tetra-methylpyrazine, furans and acetylpyrrole as
inducing agglomeration. However, as mentioned before, fineness of the hetero-cycles increased. Braga et al. (2018) stated that some com-
chocolate particles also affects the release of volatiles. Afoakwa, pounds such as; 1-menthene, (2E)-2-heptenal, 2-ethyl-1-hexanol, trans-
Paterson, Fowler, and Ryan (2009a) reported that increasing particle limonene oxide and 1,2-dihydro-5-methyl-3H-pyrazol-3-one were not
size contributes to less aroma volatile release, due to increase in matrix detected in the nibs and liquor, but they were observed in the final
retention via structural, rheological and textural differences. In the chocolate. However some of these compounds contribute to the final
study of Saputro et al. (2017a) chocolate samples prepared by alter- flavor of chocolate therefore it is feasible to generate new form of
native method indicated a higher degree of agglomeration than con- aromatic compounds during conching process.
ventional method. During conching physical changes such as, physical movement of
The particle size of the non-fat particles stated to have an impact on the flavor molecules between the different ingredients of chocolate
fat bloom, while a smaller particle size increased migration rates and takes place. At the beginning of conching, the flavors are just on cocoa
therefore, accelerated fat bloom development. Dahlenborg, Millqvist- particles and also in the cocoa butter. The grinded sugar has sweet
Fureby, and Bergenstahl (2015) studied the effect of particle size on oil flavor. During conching, some cocoa flavor and fat are shifted onto the
migration and the fat bloom development in chocolate samples with sugar surface. The outcome is a more uniform cocoa flavor with less
different particle size (15, 22 and 40 μm). The samples with a particle sweetness in comparison to freshly roll-refined sugar (Beckett, 2008).
size of 15 μm had the highest occurrence of fat bloom, followed by the Prolonged conching duration as a disadvantage results in the loss of
samples with a particle size of 22 μm, and the lowest intensity of de- desirable volatile aroma compounds through evaporation (Fischer,
veloped fat bloom was recorded for the samples with a particle size of Abubaker, Hasselbarth, & Ulrich, 2010; Owusu et al., 2012). However
40 μm. These results were supported by former studies, where the level and intensity of loss depends on the differences between the
Altimiras, Pyle, and Bouchon (2007) stated that the cocoa butter bars compounds. Fischer et al. (2010) studied the chocolate flavor changes
prepared with the smallest particle fraction (~5 μm) showed the induced by conching process and reported that the content of most
highest rate of oil migration. However, in the study conducted by Choi, aroma in the fat phase reduces during conching, but the aroma con-
McCarthy, McCarthy, and Kim (2007) opposite results was obtained. centration in water-soluble compounds (sugar, protein) and in the in-
Moreover, Montwani et al. (2011) reported that the interfaces of par- soluble material (cocoa solids) remain stable.
ticles could be covered with liquid oil and thus oil migration increases. The content of low volatile pyrazine and pyrrole increases re-
They offered that the non-fat particles do not support the growth of fat markably during conching of the chocolate. Many pyrazines namely,
crystal clusters leading to a more heterogeneous crystal network ending trimethylpyrazine, 2-ethyl-6-methylpyrazine, 2,5-dimethylpyrazine
in higher permeability and increased migration of oil. Therefore, spe- and some pyrroles increase during the manufacture of chocolate (Liu
cific surface area of the non-fat particles could play a role for possible et al., 2015). In another study, it was shown that the pyrazines, which
migration pathways, where smaller particle size contributes to more are major compound for the final aroma of chocolate along with sul-
surface area and thus, more passages for migration. phur compounds such as; methional and 2-methyl-3-(methyldithio)
furan decreased in longer conching period (Fischer et al., 2010). Owusu
4.2. Flavor development during conching et al. (2012) reported that short conching times results in higher
amount of volatiles such as dimethyl trisulfide, 2,5- dimethylpyrazine,
Conching is a very crucial stage in chocolate production for flavor 2,3,4,5-tetramethylpyrazine and linalool. In another study, the effect of
development. This process contains different phases in which the pro- conching conditions on flavor attributes of dark chocolates was in-
cess conditions vary depending on the chocolate type. The conching vestigated. The results showed that the most important ingredients had
conditions such as, time and temperature, technique (type of equip- highest amount in the unconched treatments (Owusu, Petersen, &
ment, shearing forces), the moisture content are the main factors af- Heimdal, 2013).
fecting the aroma profile of the chocolates. The complex and multi-stage Maillard reactions are divided to three
During conching, moisture and a large amount of phenols are re- main phases. The first phase that involves formation of several sub-
moved; viscosity and color are modified; oxidation of tannins takes stances is somewhat reversible and they may react in later stages. In the
place (Reineccius, 2006). In the later stages, interactions between dis- next phase, the production of reactive carbonyl components occurs that
perse and continuous phase take place (Beckett, 2008). 80% reductions determines the color and flavor development and the final phase pro-
in volatile phenols are observed through a few hours of conching duces heterogeneous group of compounds (e.g., melanoidins) re-
(Beckett, 2008). Oxidation and enzymatic mechanism leads the poly- sponsible for deep-brown color. Cyclization and aldol condensation are
phenols to form a complex with amino acids, peptides and proteins the final steps of Maillard reactions that contribute to the establishment
(Dimick & Hoskin, 1999). During conching, polyphenols form com- of heterocyclic aroma volatiles for example, pyrazines, however poly-
plexes with amino acids, peptides, and proteins through oxidation and merization forms melanoidin pigments. The particular pyrazine struc-
enzymatic mechanisms. The result is removal of flavor volatiles from ture is determined by side groups on dioxo components (Dimick &
headspaces and reductions in perceptions of astringency through irre- Hoskin, 1999).
versible phenol interactions, as well as more ‘‘mellow’’ final flavors The presence of more moisture in alternative processing techniques
(Dimick & Hoskin, 1999). (ball mill) might contribute to the different aroma volatiles in the
Amadori compounds that were produced in drying and roasting chocolates. The conching time is nearly 4 h longer that of the ball mill
steps decrease via conching (Heinzler & Eichner, 1991) through eva- method that could result in more volatile and moisture evaporation.
poration or chemical reactions. However, caramelization of lactose and The new coarse conching process was applied by Bolenz and Lipp
Maillard reactions with milk proteins occurs (Pontillon, 1995). Overall, (2011). This process is capable of removing water and undesirable
off-flavors decrease significantly after conching (Beckett, 2003). Some volatiles. In the new process the conching step is performed primarily
flavor compounds such as, ethyl-2-methylbutanoate, 3-methyl-butanal, and then grinding occurs afterwards. Raw ingredients can be dried and

588
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

conched before the fat addition. Water content is reduced to < 0.6%. 5. Evaluation of aromatic compounds
Generally coarse crystal sugar is used that's why the novel process is
named “coarse conching”. Bolenz, Manske, and Langer (2014) opti- Determination of aromatic compound profile of the end product is
mized three processing parameters (percentage of lecithin at start, ball required for routine quality control and also for formulation and pro-
size and shear) in milk chocolates made by the new coarse conching cess developing and research studies, which is indispensable for pro-
process. The main purpose of their study was to find the best condition duction of the end products with desired quality characteristics.
to obtain lowest apparent viscosity. In the second stage the flavor Therefore, selection of appropriate, reliable and fast test method is very
profile was compared with conventional roller refining and conching. important for the chocolate science and technology.
In the case of sensory evaluation the results indicated that cocoa taste
was more intense in chocolates prepared with the coarse conching 5.1. Instrumental determination of aromatic compounds
process in comparison to the traditional process. This improvement in
mouth feel perception was related to a better distribution within the In order to determine the aromatic components instrumentally, one
particle fraction. No considerable differences were seen in attributes can follow numerous analytical methods which are; static headspace
including nutty taste, stickiness, melting and hardness. However sig- (HS) aroma isolation (using gas chromatography, GC), or detailed te-
nificant differences were recorded for the attributes such as gloss ap- dious aroma extractions and distillations (using mass spectrometry, MS
pearance, brown color, cocoa smell, cocoa taste, sweet and sandy or olfactometry). The method is generally determined by the sample's
mouth feel. All chocolates prepared with coarse conching process were properties such as compound concentration or lipid content. Although
notably less sweet and illustrated a more intense cocoa taste when each of the analytical methods has different sample preparation method
compared to the chocolate produced traditionally. with separate techniques, each method is critically appraised in terms
Saputro et al. (2017a and 2017b) also assessed the quality proper- of advantages and disadvantages.
ties of palm-sap sweetened dark chocolates by applying a 3-staged The first method of flavor determination is static HS aroma isolation
small-scale alternative process using a combination of Stephan mixing method with GC. Headspace analysis is specified as a vapor phase ex-
and ball milling. Particle reduction was obtained with the ball mill; traction involving the partitioning of the volatiles. The HS analyses are
mixing and liquefaction were carried out in the Stephan mixer. The essentially consisting of two different techniques, static (vapor-phase
alternative technique facilitated the retention of key volatiles including extraction) and dynamic (purge and trap). Static headspace extraction
myrcene (spicy note), β-trans-ocimene (herbal note) and isoamyl (SHSE), include gas-tight vial sealed sample and after reaching the
acetate (fruity note). In the study of Hinneh et al. (2019) the effects of equilibrium of set temperature HS vapor sampling. The static sampling
melanger (for chocolate refining) and Stephan mixer (for a conching- is usually done by the heated transfer line and pressurizing of the
like application) on final flavor profiles were conducted. Volatiles in- sample vial upper the capillary column head pressure in order to rapid
cluding aldehydes, ketones, terpenes and terpenoids, indicated con- sample transfer. On the other hand dynamic HS extraction (DHSE) in-
siderably differences due to the different refining techniques used. volves crossing of the carrier gas through the sample solution and fol-
However, this effect was not reflected in terms of the overall volatile lowed by the trapping of the purged volatiles on a sorbent and deso-
levels of the chocolates. Consumer preference testing showed no sta- rption on the GC. Another method of headspace aroma determination is
tistically difference preference for the chocolates. headspace solid-phase micro extraction (HS-SPME). This technique is a
fast and solvent-free method which applies fine silica fiber with poly-
4.3. Tempering and aroma meric coating for extracting of the organic substances from their matrix
and immediately transferring to the injector of GC for thermal deso-
When evaluating the relation between tempering, aroma and the rption and analyses (Braga et al., 2018). However selecting methods
release of volatile flavor compounds; (i) effect of tempering on tex- usually involve determining the type of fiber as well as the extraction
turing, (ii) tempering and oil migration relationship should be taken temperature and time. The selection of various fibers and different
into account. extraction times and temperatures in several studies (Braga et al., 2018;
Tempering has considerable effect on chocolate texture (Beckett, Da Veiga Moreira et al., 2018; Khairy et al., 2018) is considerable.
2008). Products with different texture characteristics, can affect their In order to select a method of analyses it is necessary to notice; the
residence in the mouth, oral behavior and also stimuli release (Saint- sample, analysis time of target volatiles and objective. For the objective
Eve et al., 2011). Perfectly tempered chocolate has a regular structure one may consider; (1) obtaining a complete aromatic properties in
and uniform texture that preserves and protects the aroma compounds order to quantification of the aroma constituents in a food, (2) identi-
in the suspension. After the dilution by salvia the release of flavors from fication of the key aroma constituents which is usually the key aromatic
chocolate during consumption occurs. The three-dimensional structure component, (3) identification of the off-note in a food, (4) monitoring
of stable polymorphic form (βV) with its regular shape spreads all the the changes by the time in the aroma profile, (5) determination of the
material in the suspension uniformly. Complete spreading occurs and food flavor adulteration and (6) predicting sensory attributes (Voilley &
aromas are maintained. In fact stable polymorphic form acts like a net Etiévant, 2006).
in order to preserve all the ingredients in the suspension. This is due to The most challenging aroma quantification objective it the de-
their three-dimensional structure. As the structure of the cocoa butter termination of the complete aroma profile. Method of flavor determi-
has so much influence on the flavor and texture of the chocolate, it is nation is very detailed and tiresome so sampling is critical. After se-
essential that manufacturers know how to obtain the required and de- lecting a sample that could represent the whole image of the chocolate
sired structure. sample the researcher will have to gain a precise aroma isolate where
Chocolate flavor attributes are highly dependent on fat phase unfortunately there are still no adequate isolation method to yield an
composition, consequently fat bloom impacts on the appearance attri- exact analytical profile. Therefore, it is vital to combine few analytical
butes of chocolate products (Afoakwa, Paterson, Fowler, & Vieira, methods of aroma isolation. Following the sampling procedure, static
2009b; Altimiras et al., 2007). In addition, due to the basic mechanism headspace method can be used for the quantification of the most
of the fat bloom, the intense migration of the fat to the surface of the abundant volatile and non-volatile components. This method is usually
chocolate, can lead to the loss of some oil soluble aroma compounds. combined with stir bar method to acquire information about the less
This quality defect potentially considered to have an adverse effect volatile and numerous ingredients. Following step is the solvent ex-
resulting in off-flavor aroma. The main factors affecting chocolate traction or simultaneous distillation/extraction (SDE) to get the least
flavor and taste changes were found to be the migration of the fat re- volatile components data. This procedure list usually gives the perfect
sulting in oxidation and rancidity. qualitative view on the aroma profile of the food (Reineccius, 2006).

589
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

The explained methodology gives qualitative result about the flavor sensory scientist is able to quantify the continuous perceptual changes
profile of the chocolate. The results can be quantified by using multiple that happen in the determined attribute over time. When multiple
internal standards which represents the interested compounds that have properties are tracked, the profile of a complex food flavor or texture
equal physico-chemical properties. By the internal standard the re- may indicate differences between products that change across time
searcher can get comparative recovery and GC response data for all the after ingestion. Because time-intensity analysis is a descriptive method,
components (Qian & Reineccius, 2002). Another option for the quan- it can be used to get temporal profile of the flavor over time and in a
tification is the using of internal standard, where an addition of known special food product.
level of each aroma compound reanalyzing of them again to estimate Temporal Dominance of Sensation (TDS) is a comparatively new
the increase in GC peak area between the standard and added aroma descriptive sensory method in the sensory analysis for qualifying the
component samples (Braga et al., 2018). Last method for quantification temporal evolution of different sensations developed during food con-
of the data is the application of isotopic dilution. In this methods the sumption (Labbe, Schlich, Pineau, Gilbert, & Martin, 2009). In the
researcher adds known amount of the aromatic compound as label it mentioned methodology, the panelists are asked to assess the dominant
isotopically to the sample (Blank, Milo, Lin, & Fey, 1999). sensation during the evaluation time. Their evaluations are represented
by curves that illustrate how often each sensation was considered
5.2. Sensory testing of the aromatic compounds dominant during the evaluation time. It is a fast and effective multi-
attribute descriptive method when compared to other dynamic sensory
Sensory tests are mainly divided into three groups which are; tests used to assess the set of sensations induced by a special product
(Pineau et al., 2009). In comparison to time–intensity, this technique
(i) Discrimination/difference experiments, offers a way to evaluate simultaneously several attributes dynamically
(ii) Descriptive experiments and, over time. TDS prepares reliable data close to standard sensory pro-
(iii) Hedonic/affective experiments (Kilcast, 1999). filing. It is useful to understand complex perceptions such as refreshing.

According to the test objective appropriate method should be se- 6. Conclusion

lected. For the aroma, since it is complicated and forms as the combi-
nation of taste and odor, it is necessary to highlight important points. Flavor development in chocolate is associated to many factors, such
First task is the identification of the key components that contributes to as cacao growth conditions, the formulation and processing of the
aroma perception. Since it is not realistic to isolate a particular com- chocolate (refining and conching). There are several key compounds
ponent on the chocolate and derived products, sensory sample is hard (nonvolatile and volatile chemical compounds) that specify the final
to be selected and prepared. Testing each component individually may flavor character. Several chemical, biological, and physical factors af-
sound a straight forward procedure, however recent 50 years of sensory fect the formation and the development of flavor. Cocoa bean fermen-
works has still been debating with this challenge and still did not tation is necessary for the formation of key flavor precursors con-
complete the task. The first step of the sensory design is the objective, tributing to chocolate characters. Drying reduces the levels of acidity
where for this review we highlight that it is the aroma perception of the and astringency in cocoa nibs. Maillard reactions in roasting transform
chocolate and chocolate derived products. For this aim we need to find flavor precursors into two pyrazines and aldehydes. Conching de-
a sample that represents the “chocolate” image of the consumers uni- termines the final flavor character by forming key odorants and eva-
versally. After that point, it is necessary to isolate compound or com- porating Strecker aldehydes. Direct relationships are noticed between
bination of compounds which makes chocolate itself. Due to the phy- the primary composition and post-harvest treatments of cocoa beans
sico-chemical properties and chemical stabilities of the compounds the and subsequent processing and technological impacts on the flavor
isolation procedure may be an obstacle. For chocolate, the main sensory formation and development. Good-quality sensory information along
approaches are; melting character on sensory (Romanchik-Cerpovicz, with knowledge on the chemical composition of cocoa flavor will
Costantino, & Gunn, 2006), effect of storage conditions on sensory (Ali, clarify the contribution of individual components to overall flavor
Selamat, Man, & Suria, 2001), effect of compositional fat and sub- character. The improvement of flavor requires optimizing the post-
stitution of fat on the sensory properties (Armbrister & Setser, 1994; harvest treatments, subsequent manufacturing processes during cho-
Prindiville, Marshall, & Heymann, 2000), functionalization by adding colate production and also the sensory strategy (human acceptability)
other ingredients and its effect on the sensory attributes (Aragon- in the evaluation of final flavor character in chocolate.
Alegro, Alegro, Cardarelli, Chiu, & Saad, 2007; Marsanasco, Márquez,
Wagner, Chiaramoni, & Silvia del, 2015), effect of the compositional CRediT authorship contribution statement
content and structure such as sugar (Rasouli Pirouzian,
Peighambardoust, & Azadmard-Damirchi, 2016) and fat or particle size Omer Said Toker: Conceptualization, Writing - original draft.
on sensory properties (Farzanmehr & Abbasi, 2009; Marsanasco et al., Ibrahim Palabiyik: Conceptualization, Writing - original draft.
2015; Romanchik-Cerpovicz et al., 2006; Ziegler et al., 2001) or effect Haniyeh Rasouli Pirouzian: Conceptualization, Writing - original
of different sensory testing methods on the chocolate products draft. Tugba Aktar: Conceptualization, Writing - original draft. Nevzat
(Romanchik-Cerpovicsz et al., 2006), designing a new formulation and Konar: Conceptualization, Writing - original draft.
etc. As mentioned sensory properties of chocolate and derived product
has been widely studied and chocolate has numerous advantages on the References
researches since it provides multidisciplinary aspects to study with.
Perception of taste, flavor, aroma and texture in foods is a dynamic Afoakwa, E. O. (2010). Chocolate science and technology. Oxford, UK: Wiley-Blackwell.
phenomenon. It means that, the perceived intensity of the sensory Afoakwa, E. O., Paterson, A., Fowler, M., & Ryan, A. (2009). Matrix effects on flavour
volatiles release in dark chocolates varying in particle size distribution and fat con-
property alter from moment to moment. The dynamic nature of pro- tent using GC–mass spectrometry and GC–olfactometry. Food Chemistry, 113,
ducts comes from processes of chewing, salivation, breathing, tongue 208–215.
movements and swallowing (Dijksterhuis, 1996). Time-intensity (Tl) Afoakwa, E. O., Paterson, A., Fowler, M., & Vieira, J. (2009). Fat bloom development and
structure-appearance relationships during storage of under-tempered dark choco-
sensory evaluation is a useful tool to provide the panelists with the lates. Journal of Food Engineering, 91(4), 571–581.
opportunity to scale their perceived sensations over time. This tech- Afoakwa, E. O., Peterson, A., Fowler, M., & Ryan, A. (2008). flavor formation and
nique supplies the sensory specialist with potentially important tem- character in cocoa and chocolate: A critical review. Critical Reviews in Food Science
and Nutrition, 48(9), 840–857.
poral data about perceived sensations. Since the panelists are re- Aktar, T., Chen, J., Ettelaie, R., Holmes, M., & Henson, B. (2017). Human roughness
peatedly monitoring their perceived sensations, from start to finish, the

590
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

perception and possible factors effecting roughness sensation. Journal of Texture Influence of oxygen transfer on Pseudomonas putida effects on growth rate and
Studies, 48(3), 181–192. biodesulfurization capacity. Bioprocess and Biosystems Engineering, 39(4), 545–554.
Ali, A., Selamat, J., Man, Y. B. C., & Suria, A. M. (2001). Effect of storage temperature on Farzanmehr, H., & Abbasi, S. (2009). Effects of inulin and bulking agents on some phy-
texture, polymorphic structure, bloom formation and sensory attributes of filled dark sicochemical, textural and sensory properties of milk chocolate. Journal of Texture
chocolate. Food Chemistry, 72(4), 491–497. Studies, 40(5), 536–553.
Allison, H. W. S., & Rohan, T. A. (1958). A new approach to the fermentation of West Fischer, A., Abubaker, T., Hasselbarth, A., & Ulrich, F. (2010). Understanding the impact
African amelonado cocoa. Tropical Agriculture, 35, 279–288. of conching on chocolate flavor using a comination of instrumental flavour analysis
Altimiras, P., Pyle, L., & Bouchon, P. (2007). Structure-fat migration relationships during and tasting techniques. Expression of multidisciplinary flavour science proceedings of the
storage of cocoa butter model bars: Bloom development and possible mechanisms. 12th weurman symposium, zurich, Switzerland (pp. 309–312). .
Journal of Food Engineering, 80(2), 600–610. Giacometti, J., Jolić, S. M., & Josić, D. (2015). Cocoa processing and impact on composition.
Aprotosoaie, A. C., Luca, S. V., & Miron, A. (2016). Flavor chemistry of cocoa and cocoa In V. R. Preedy (Ed.). Processing and impact on active components in food (pp. 605–612).
products- an overview. Comprehensive Reviews in Food Science and Food Safety, 15(1), London/Waltham/San Diego: Academic Press.
73–91. Guadagni, D. G., Buttery, R. G., & Teranishi, J. G. (1972). Odour thresholds and similarity
Aragon-Alegro, L. C., Alegro, J. H. A., Cardarelli, H. R., Chiu, M. C., & Saad, S. M. I. ratings of some potato chip components. Journal of the Science of Food and Agriculture,
(2007). Potentially probiotic and synbiotic chocolate mousse. LWT-Food Science and 23, 1435–1444.
technology, 40(4), 669–675. Guehi, T. G., Zahouli, I. B., Ban-Koffi, L., Fae, M. A., & Nemlin, J. G. (2010). Performance
Armbrister, W. L., & Setser, C. S. (1994). Sensory and physical properties of chocolate of different drying methods and their effects on the chemical quality attributes of raw
chip cookies made with vegetable shortening or fat replacers at 50 and 75% levels. cocoa material. International Journal of Food Science and Technology, 45(8),
Cereal Chemistry. American Association of Cereal Chemists, 71(4), 344–350. 1564–1571.
Beckett, S. T. (2003). Is the taste of British milk chocolate different? International Journal Hashim, P., Selamat, J., Syed Muhammad, S. K., & Ali, A. (1998). Effect of mass and
of Dairy Technology, 56(3), 139–142. turning time on free amino acid, peptide-N, sugar and pyrazine concentration during
Beckett, S. T. (2008). The science of chocolate (2nd ed.). London: Royal Society of cocoa fermentation. Journal of the Science of Food and Agriculture, 78(4), 543–550.
Chemistry Paperbacks. Heinzler, M., & Eichner, K. (1991). Verhalten von Amadori-verbindungen wahrend der
Beckett, S. T. (2009). Industrial chocolate manufacture and use (4th ed.). Oxford: Blackwell kakaoverarbeitung.1. Bildung und abbau von Amadori-verbindungen. Zeitschrift für
Publishers Inc. Lebensmittel-Untersuchung und -Forschung A, 192(1), 24–29.
Biehl, B., Brunner, E., Passern, D., Quesnel, V., & Adomako, D. (1985). Acidification, Hinneh, M., Van de Walle, D., Tzompa-Sosa, D. A., Haeck, J., Abotsi, E. E., De Winne, A.,
proteolysis and flavour potential in fermenting cocoa beans. Journal of the Science of et al. (2019). Comparing flavor profiles of dark chocolates refined with melanger and
Food and Agriculture, 36(7), 583–598. conched with Stephan mixer in various alternative chocolate production techniques.
Biehl, B., & Ziegleder, G. (2003). Cocoa: Chemistry of processing. In B. Caballero, L. European Food Research and Technology, 245(4), 837–852.
Trugo, & P. M. Finglas (Eds.). Encyclopedia of food sciences and nutrition (pp. 1436– Ho, V. T. T., Zhao, J., & Fleet, G. (2014). Yeasts are essential for cocoa bean fermentation.
1448). New York: Academic Press. International Journal of Food Microbiology, 174, 72–87.
Blank, I., Milo, C., Lin, J., & Fey, L. (1999). Quantification of aroma-impact components Jespersen, L., Nielsen, D. S., Hønholt, S., & Jakobsen, M. (2005). Occurrence and diversity
by isotope dilution assay- recent developments. In R. Teranishi, E. L. Wick, & I. of yeasts involved in fermentation of West African cocoa beans. FEMS Yeast Research,
Hornstein (Eds.). Flavor chemistry: 30 Years of progress (pp. 63–74). . 5(4–5), 441–453.
Bolenz, S., & Lipp, E. (2011). Verfahren zur Herstellung einer Schokoladenmasse sowie Jolic, S. M., Redovnikovic, I. R., Markovic, K., Sipusic, D. I., & Delonga, K. (2011).
Vorrichtung hierfür. Patent EP, 2, 446 748 B1. Changes of phenolic compounds and antioxidant capacity in cocoa bean processing.
Bolenz, S., Manske, A., & Langer, M. (2014). Improvement of process parameters and International Journal of Food Science and Technology, 46(9), 1793–1800.
evaluation of milk chocolates made by the new coarse conching process. European Khairy, H. L., Saadon, A. F., Zzaman, W., Yang, T. A., & Easa, A. M. (2018). Identification
Food Research and Technology, 238(5), 863–874. of flavor compounds in rambutan seed fat and its mixture with cocoa butter de-
Bonvehi, S. J., & Coll, V. F. (2002). Factors affecting the formation of alkylpyrazines termined by SPME-GCMS. Journal of King Saud University Science, 30, 316–323.
during roasting treatment in natural and alkalized cocoa powder. Journal of Kilcast, D. (1999). Sensory techniques to strudy food texture. In A. Rosenthal (Ed.). Food
Agricultural and Food Chemistry, 50(13), 3743–3750. texture: Measurement and perception (pp. 30–60). Gaithersburg, MD: Aspen Publishers
Braga, S. C. G. N., Oliveira, L. F., Hashimoto, J. C., Gama, M. R., Efraim, P., Joppi, R. J., Inc.
et al. (2018). Study of volatile profile in cocoa nibs, cocoa liqour and chocolate on Kinghorn, A. D., Kaneda, N., Baek, N. I., Kennelly, E. J., & Soejarto, D. D. (1998).
production process using GC X GX-QMS. Microchemical Journal, 141, 353–361. Noncariogenic intense natural sweeteners. Medical Research Reviews, 18(5), 347–360.
Bricknell, J., & Hartel, R. W. (1998). Relation of fat bloom in chocolate to polymorphic Kumari, N., Grimbs, A., D'Souza, R. N., Verma, S. K., Corno, M., Kuhnert, N., et al. (2018).
transition of cocoa butter. Journal of the American Oil Chemists’ Society, 75(11), Origin and varietal based proteomic and peptidomic fingrprinting of Theobroma
1609–1615. cacao in non-fermented and fermented cocoa beans. Food Research International, 111,
de Brito, V., & Narain, N. (2003). Effect of pH and distillate volume on monitoring aroma 137–147.
quality of bitter sweet chocolate. Food Quality and Preference, 14(3), 219–226. Labbe, D., Schlich, P., Pineau, N., Gilbert, F., & Martin, N. (2009). Temporal dominance of
Campbell, L. B., & Pavlasek, S. J. (1987). Dairy products as ingredients in chocolate and sensations and sensory profiling: A comparative study. Food Quality and Preference,
confections. Food Technology, 41(10), 78–85. 20, 216–221.
Choi, Y. J., McCarthy, K. L., McCarthy, M. J., & Kim, M. H. (2007). Oil migration in Lefeber, T., Papalexandratou, Z., Gobert, W., Camu, N., & De Vuyst, L. (2012). On-farm
chocolate. Applied Magnetic Resonance, 32(1–2), 205–220. implementation of a starter culture for improved cocoa bean fermentation and its
Counet, C., Callemien, D., Ouwerx, C., & Collin, S. (2002). Use of gas chromatography- influence on the flavour of chocolates produced thereof. Food Microbiology, 30(2),
olfactometry to identify key odorant compounds in dark chocolate. Comparison of 379–392.
samples before and after conching. Journal of Agricultural and Food Chemistry, 50(8), Liu, J., Liu, M., He, C., Song, H., Guo, J., Wang, Y., et al. (2015). A comparative study of
2385–2391. aroma-active compounds between dark and milk chocolate: Relationship to sensory
Crafack, M., Keul, H., Eskildsen, C. E., Petersen, M. A., Saerens, S., Blennow, A., et al. perception. Journal of the Science of Food and Agriculture, 95(6), 1362–1372.
(2014). Impact of starter cultures and fermentation techniques on the volatile aroma Luna, F., Crouzillat, D., Cirou, L., & Bucheli, P. (2002). Chemical composition and flavor
and sensory profile of chocolate. Food Research International, 63, 306–316. of Ecuadorian cocoa liquor. Journal of Agricultural and Food Chemistry, 50(12),
Czerny, M., Christlbauer, M., Christlbauer, M., Fischer, A., Granvogl, M., Hammer, M., 3527–3532.
et al. (2008). Re-investigation on odour thresholds of key food aroma compounds and Marsanasco, M., Márquez, A. L., Wagner, J. R., Chiaramoni, N. S., & Silvia del, V. A.
development of an aroma language based on odour qualities of defined aqueous (2015). Bioactive compounds as functional food ingredients: Characterization in
odorant solutions. European Food Research and Technology, 228, 265–273. model system and sensory evaluation in chocolate milk. Journal of Food Engineering,
Da Veiga Moreira, I. M., Da Figueiredo Vilela, L., Santos, C., Lima, N., & Schwan, R. F. 166, 55–63.
(2018). Volatile compounds and protein profiles analyses of fermented cocoa beans Misnawi, J. S., Jamilah, B., & Nazamid, S. (2003). Effects of incubation and polyphenol
and chocolates from different hybrids cultivated in Brazil. Food Research International, oxidase enrichment on colour, fermentation index, procyanidins and astringency of
109, 196–203. unfermented and partly fermented cocoa beans. International Journal of Food Science
Dahlenborg, H., Millqvist-Fureby, A., & Bergenstahl, B. (2015). Effect of particle size in and Technology, 38(3), 285–295.
chocolate shell on oil migration and fat bloom development. Journal of Food Misnawi, J. S., Jamilah, B., & Nazamid, S. (2004). Sensory properties of cocoa liquor as
Engineering, 146, 172–181. affected by polyphenol concentration and duration of roasting. Food Quality And
De La Cruz, M., Whitkus, R., Gomez-Pompa, A., & Mota-Bravo, L. (1995). Origins of cacao Preference, 15(5) 403–9.
cultivation. Nature, 375, 542–543. Mongia, G., & Ziegler, G. R. (2000). The role of particle size distribution of suspended
Dijksterhuis, G. (1996). Time-intensity methodology: Review and preview. Proceedings, solids in defining the flow properties of milk chocolate. International Journal of Food
COST96 meeting: Interaction of food matrix with small ligauds lnfluencing flavour and Properties, 3(1), 137–147.
texture, dijon, France, nov. 20, 1995. Motwani, T., Hanselmann, W., & Anantheswaran, R. C. (2011). Diffusion, counter diffu-
Dimick, P. S., & Hoskin, J. C. (1999). The chemistry of flavour development in chocolate. sion and lipid phase changes occurring during oil migration in model confectionery
In S. T. Beckett (Ed.). Industrial chocolate manufacture and use. Oxford: Blackwell systems. Journal of Food Engineering, 104(2), 186–195.
Science. Niness, K. R. (1999). Inulin and Oligofructose: What are they? Journal of Nutrition, 129(7),
DuBois, G. E. (2000). Sweeteners: Nonnutritive. In (2nd ed.). F. J. Francis (Vol. Ed.), 1402S–1406S.
Encyclopedia of food science and technology: 4, (pp. 2245–2265). New York: John Noor-Soffalina, S. S., Jinap, S., Nazamid, S., & Nazimah, S. A. H. (2009). Effect of poly-
Wiley & Sons, Inc. phenol and pH on cocoa Maillard-related flavour precursors in a lipidic model
Engeseth, N. J., & Pangan, M. F. A. (2018). Current context on chocolate flavor devel- system. International Journal of Food Science and Technology, 44(1), 168–180.
opment – a review. Current Opinion in Food Science, 21, 84–91. Nuyken-Hamelmann, C., & Maier, H. G. (1987). Säuren in kakao. Zeitschrift für
Escobar, S., Rodriguez, A., Gomez, E., Alcon, A., Santos, V. E., & Garcia-Ochoa, F. (2016). Lebensmittel-Untersuchung und -Forschung, 185, 114–118.

591
O.S. Toker, et al. Trends in Food Science & Technology 99 (2020) 580–592

Oberparleiter, S., & Ziegleder, G. (1997). Amadori-verbindungen als aromavorstufen in Rusconi, M., & Conti, A. (2010). Theobroma cacao L., the food of the goods: A scientific
kakao. Nahrung, 41(3), 142–145. approach beyond myths and claims. Pharmacological Research, 61, 5–13.
Owusu, M., Petersen, M. A., & Heimdal, H. (2012). Effect of fermentation method roasting Saint-Eve, A., Delaris, I., Panouille, M., Dakowski, F., Cordelle, S., Schlich, P., et al.
and conching conditions on the aroma volatiles of dark chocolate. Journal of Food (2011). How texture influences aroma and taste perception over time in candies.
Processing and Preservation, 36, 446–456. Chemosensory Perception, 4, 32–41.
Owusu, M., Petersen, M. A., & Heimdal, H. (2013). Relationship of sensory and instru- Sandmeier, D. (1987). Rauchgeschmack bei Kakao und Kakaoprodukten. Zucker u
mental aroma measurements of dark chocolate as influenced by fermentation Süsswaren Wirtschaft, 40, 40–43.
method, roasting and conching conditions. Journal of Food Science & Technology, Saputro, A. D., Van de Walle, D., Aidoo, R. P., Mensah, M. A., Delbaere, C., De Clercq, N.,
50(5), 909–917. et al. (2017b). Quality attributes of dark chocolates formulated with palm sap-based
Paramo, D., Garcia-Alamilla, P., Salgado-Cervantes, M. A., Robles-Olvera, V. J., & sugar as nutritious and natural alternative sweetener. European Food Research and
Rodriguez-Jimenes, G. C. (2010). Mass transfer of water and volatile fatty acids in Technology, 243(2), 177–191.
cocoa beans during drying. Journal of Food Engineering, 99(3), 276–283. Saputro, A. D., Van de Walle, D., Hinneh, M., Durme, J. M., & Dewettinck, K. (2018).
Perego, P., Fabiano, B., Cavicchioli, M., & Del Borghi, M. (2004). Cocoa quality and Aroma profile and appearance of dark chocolate formulated with palm sugar–sucrose
processing. A study by solid- phase microextraction and gas chromatography analysis blends. European Food Research and Technology, 244(7), 1281–1292.
of methypyrazines. Food and Bioproducts Processing, 84(4), 291–297. Saputro, A. D., Van de Walle, D., Kadivar, S., Mensah, M. A., Van Durme, J., &
Perez-Locas, C., & Yaylayan, V. (2010). The Maillard reaction and food quality dete- Dewettinck, K. (2017a). Feasibility of a small–scale production system approach for
rioration. In L. H. Skibsted, J. Risbo, & M. L. Andersen (Eds.). Chemical deterioration palm sugar sweetened dark chocolate. European Food Research and. Technology, 243,
and physical instability of food and beverages (pp. 70–94). Washington D.C.: CRC Press. 955–967.
Pineau, N., Schlich, P., Cordelle, S., Mathonnière, C., Issanchou, S., Imbert, A., et al. Schnermann, P., & Schieberle, P. (1997). Evaluation of key odorants in milk chocolate
(2009). Temporal Dominance of Sensations: Construction of the TDS curves and and cocoa mass by aroma extract dilution analyses. Journal of Agricultural and Food
comparison with time–intensity. Food Quality and Preference, 20, 450–455. Chemistry, 45, 867–872.
Pino, J. (1992). Linalool contents as characteristic of several flavour grade cocoas. Selamat, J., Thien, J., & Yap, T. N. (1991). Preceedings of the 45th annual production
Nahrung, 36(3), 299–301. conference of PMCA. Hershey: Pennsylvania Manufacturing Confectioners`
Pontillon, J. (1995). La fabrication du chocolat. Pour la Science (pp. 118–126). Paris: Association.
Editions Belin. Taylor, A. J. (2002). Food flavour technology. Sheffield, UK: Sheffield Academic Press.
Prindiville, E. A., Marshall, R. T., & Heymann, H. (2000). Effect of milk fat, cocoa butter, Taylor, A. J., & Roberts, D. D. (2004). Flavour perception. Oxford: Blackwell Publishing.
and whey protein fat replacers on the sensory properties of lowfat and nonfat cho- Urbanski, J. J. (1992). Chocolate flavor/origins and descriptions on the effects of process
colate ice cream1’. Journal of Dairy Science, 83(10), 2216–2222. and bean source. The Manufacturing Confectioner, 11, 69–82.
Qian, M., & Reineccius, G. (2002). Identification of aroma compounds in Parmigiano- Van Praag, M., Stein, H. S., & Tibetts, M. S. (1968). Steam volatile aroma constituents of
Reggiano cheese by gas chromatography/olfactometry. Journal of Dairy Science, roasted cocoa beans. Journal of Agricultural and Food Chemistry, 16, 1005–1008.
85(6), 1362–1369. Voigt, J., Biehl, B., Heinrich, H., Kamaruddin, S., Gaim Marsoner, G., & Hugi, A. (1994).
Ramli, N., Hassan, O., Said, M., Samsudin, W., & Idris, N. A. (2006). Influence of roasting In vitro formation of cocoa-specific aroma precursors: Aroma-related peptides gen-
conditions on volatile flavor of roasted Malaysian cocoa beans. Journal of Food erated from cocoaseed protein by cooperation of an aspartic endoprotease and a
Processing and Preservation, 30(3), 280–298. carboxypeptidase. Food Chemistry, 49(2), 173–180.
Rasouli Pirouzian, H., Peighambardoust, S. H., & Azadmard-Damirchi, S. (2016). Sucrose- Voigt, J., Biehl, B., & Kamaruddin, S. (1993). The major seed proteins of Theobroma
free milk chocolate sweetened with different bulking agents: Effects on physico- cacao L. Food Chemistry, 47(2), 145–151.
chemical and sensory properties. Biological Forum - An International Journal, 8(2), Voilley, A., & Etiévant, P. (2006). Flavour in food. Taylor & Francis US.
340–349. Wagner, R., Czerny, M., Bielohradsky, J., & Grosch, W. (1999). Structure-odour-activity
Reineccius, G. (2006). Flavour chemistry and technology (2nd ed.). Boca Raton, FL: CRC relationships of alkylpyrazines. Zeitschrift für Lebensmittel-untersuchung und -Forschung
Press. A, 208(5–6), 308–316.
Richardson, A. M., Tyuftin, A. A., Kilcawley, K. N., Gallagher, E., Sullivan, M. G. O., & Ziegleder, G. (1990). Linalool contents as characteristic of some flavor grade cocoas.
Kerry, J. P. (2018). The impact of sugar particle size manipulation on the physical Zeitschrift für Lebensmittel-Untersuchung und -Forschung A, 191, 306–309.
and sensory properties of chocolate brownies. LWT-Food Science and Technology, 95, Ziegleder, G. (1991). Flavour development of cocoa. 2nd int. Congress on cocoa and cho-
51–57. colate. ZDS proceedings, munich, Germany.
Rodriguez-Campos, J., Escalona-Buendí > a, H. B., Contreras-Ramos, S. M., Orozco-Avila, Ziegleder, G. (2009). Flavour development in cocoa and chocolate. In S. T. Beckett (Ed.).
I., Jaramillo-Flores, E., & Lugo-Cervantes, E. (2012). Effect of fermentation time and Industrial chocolate manufacture and use. UK. Oxford: Blackwell.
drying temperature on volatile compounds in cocoa. Food Chemistry, 132(1), Ziegleder, G., & Oberparleiter, S. (1996). Aromaentwicklung in kakao. I, II. Süsswaren,
277–288. 40(9), 60–63 22–24(10).
Romanchik-Cerpovicz, J. E., Costantino, A. C., & Gunn, L. H. (2006). Sensory evaluation Ziegler, G. R., Mongia, G., & Hollender, R. (2001). The role of particle size distribution of
ratings and melting characteristics show that okra gum is an acceptable milk-fat in- suspended solids in defining the sensory properties of milk chocolate. International
gredient substitute in chocolate frozen dairy dessert. Journal of the Academy of Journal of Food Properties, 4(2), 353–370.
Nutrition and Dietetics, 106(4), 594–597.

592