International Journal of Food Science and Technology 2007, 42, 319329

319

Original article
Elaboration of a fruit wine from cocoa (Theobroma cacao L.) pulp
Disney Ribeiro Dias,1 Rosane Freitas Schwan,1* Evandro Sena Freire2 & Rogerio dos Santos Serodio3
1 Biology Department, Federal University of Lavras, Cx. Postal 37, 37.200-000, Lavras, MG, Brazil
2 Departamento de Ciencias Exatas e Tecnologicas, Universidade Estadual de Santa Cruz, Km 16 da Rodovia Ilheus/Itabuna, BA. 45650-000,
Ilheus, BA, Brazil
3 Cocoa Research Centre, CEPLAC, Km 22 Rodovia Ilheus, Itabuna, 45.600-000, Itabuna, BA, Brazil
(Received 8 August 2005; Accepted in revised form 20 March 2006)

Summary

The objectives of this study were the selection of a strain of Saccharomyces cerevisiae, the elaboration of a
fermentative process using cocoa (Theobroma cacao L.) fruit pulp, and the assessment of the acceptance of
the elaborated beverage. Three S. cerevisiae strains (CA116, CA1162 and CA1183) were assessed while
growing in a fruit pulp medium at dierent temperatures. The ethanol:biomass and glycerol:biomass ratios
showed that there were no signicant dierences among the three strains at dierent temperatures. However,
the strain CA1183 reached a higher ethanol production and yield and it was chosen as a starter to produce
the cocoa beverage. The concentration of higher alcohols, methanol, esters and acetaldehyde found in the
elaborated beverage was in accordance with the standards established for table wine. Sensory analysis
revealed a high degree of acceptance amongst the great majority of tasters. It can be concluded that pulp
processing into an alcoholic beverage is a realistic additional way of utilisation of the cocoa fruit.

Keywords

Alcoholic fermentation, cocoa, fruit wine, Saccharomyces, Theobroma cacao.

Introduction

Cocoa (Theobroma cacao L.) is world-wide known for its

beans used in the manufacture of chocolate. For a long
time the production and commercialisation of cocoa has
been the basis for the economy of some Brazilian states,
mainly Bahia. From 1990, because of the nancial crisis
in the international market of cocoa and the emergence of
the witches broom disease caused by the pernicious
fungus Crinipellis perniciosa in Bahia (Brazil), the agroindustry of cocoa declined quickly, causing a decrease in
production and exportation of cocoa beans. In this
period producers started to use and commercialise the
pulp of the fruit as an income source, which was almost
completely discarded at the height of the exportation of
beans. This alternative is one of the current forms of
commercialisation of the cocoa in Brazil. The cocoa pulp
is a substrate rich in nutrients, which can be used in
industrial processes for by-product manufacture (Freire
et al., 1999; Schwan & Wheals, 2004).
*Correspondent: Fax: + 55 (35) 3829 1100;
e-mail: [email protected]
Present address: Disney Ribeiro Dias, Unilavras Centro Universitario de Lavras, Rua Pe. Jose Poggel 506, 37.200-000, Lavras, MG,
Brazil.

Cocoa pulp can be readily fermented by yeasts such as

Saccharomyces cerevisiae, producing an alcoholic beverage. Saccharomyces cerevisiae has been used in
fermentative processes for thousands of years, according
to the rst historical stories of the production of beer
and wine (Rose, 1977; Hansen & Kielland-Brandit,
1996; Demain, 2000; Ostergaard et al., 2000). Because of
commercial importance of this microorganism, strains
with good fermentative characteristics have been selected and commercialised in dehydrated form and/or
lyophilised to be employed in breweries, wineries and
other industries (Fleet & Heard, 1993; Colagrande et al.,
1994; Ubeda-Iranzo et al., 2000; Cappello et al., 2004).
A range of environmental factors inuences the production of metabolites and survival of yeasts during
industrial fermentations. The main factors are temperature, pH, sugars concentration and acidity of fruit juice
(substrate). In case of yeasts, temperature and tolerance
of ethanol have an important inuence on their
performance (Heard & Fleet, 1988; Walker, 1998).
Winemaking is a good example of the biotechnological
evolution in the beverage production, passing from artto science-based technology. As the grape, several other
fruits can be used for forming musts that can be
submitted to alcoholic fermentation. Experiments with
fruits for fermented beverage production have been
carried out in the last two decades, using musts of

doi:10.1111/j.1365-2621.2006.01226.x
 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

320

Elaboration of a fruit wine from cocoa D. R. Dias et al.

banana, coconut water, pineapple and tomato (Garcia

et al., 1976), mango (Akubor, 1996), pineapple (Ayogu,
1999), pupunha (Oliveira et al., 2001), orange (Selli
et al., 2003), yellow mombin (Dias et al., 2003) and
carambola (Vidya & Neela, 2004). However, because of
the great dierence of composition between the fruits,
there is a necessity for more studies for the production
of these beverages notably about the strain to be used,
the ideal temperature of fermentation and the type of
treatment that has to be applied to the must of the fruit,
or the proper fruit, in the pre-fermentative phase and
during the fermentation.
World-wide Brazil is one of the countries with a large
production of fruits and it is also one of the biggest
exporting countries. In relation to the world-wide
production of cocoa, the country occupies the fth
position (Schwan & Wheals, 2004; FAOSTAT Data,
2005) after Ivory Coast, Ghana, Indonesia and Nigeria.
However, not all of the harvest nds its destination in
the market, because of overproduction or damage. New
forms of use or processing of the production excess must
be developed, in order to prevent losses and to generate
prots. One of the alternatives is its use in fermented
beverage production.
The purpose of this study was to use cocoa pulp for
the production of a fermented alcoholic beverage. This
was carried out under three steps: selection of strains of
S. cerevisiae, elaboration of a ow diagram for the
process of production and evaluation of the chemical
and sensory properties of the nal beverage.

Viability determination

For viability measurements, 100 lL of appropriate

dilutions of the cultures were plated (in triplicate) on
YPD plates. Plates were incubated at 30 C until the
appearance of colonies (13 days), and the number of
colony-forming units per millilitre of cell culture was
determined. Microscopic cell viability was measured
using methylene blue.
Budding index

Budding index, dened as the fraction of all cells with

visible buds, was measured for about 300 cells using a
microscope. This measure has been shown to be a good
index for determining the rate of cell proliferation (Lord
& Wheals, 1980).

Materials and methods

Yeast strains

The three S. cerevisiae strains have previously been

isolated from natural cocoa and sugar cane fermentations and selected because of their dierences in
fermentative behaviour in synthetic fermentation medium (Schwan, 1988; Schwan et al., 2001). The strains
belong to the Microbial Physiology laboratory at DBI/
UFLA and are codied as CA116, CA1162 and
CA1183. These strains were cultured for 48 h at 30 C
on fresh YPD medium (dextrose 20 g L)1; bacteriological peptone 20 g L)1; yeast extract 10 g L)1) solidied
with 2% agar when required, before they were used in
the fermentations.
Fermentation experiments

Batch fermentations were carried out in 500 mL

Erlenmeyer asks containing 450 mL of fruit fermentation medium (FFM) containing glucose 10.0 g L)1;
KH2PO4 4.5 g L)1; (NH4)2SO4 3.0 g L)1; yeast extract
1.0 g L)1; MgSO47H2O 0.25 g L)1 and CaCl2
0.25 g L)1; at pH 5.0 and cocoa pulp in order to obtain

International Journal of Food Science and Technology 2007

a nal sugar concentration of 200 g L)1 and covered

with sterile hydrophobic cotton cap. The initial yeast
inoculum was 1.0 106 cells mL)1 from FFM overnight
cultures. Cultures were evaluated about their fermentative ability in three dierent temperatures of incubation:
18, 22 and 25 C without shaking. At several time points
during the fermentation process, bottles were gently
shaken and samples were taken for optical density
(OD600) and viability measurements, determination of
residual sugars (Miller, 1959) and chemical analysis.
Each batch fermentation was carried out at least four
times. After 24 h or when no residual sugars were left,
the supernatant culture medium (FFM) was descarted
and a fresh culture medium was added to the cells. This
procedure was carried out three times in a row. Sugar
consumption was plotted against time. The maximum
fermentation rate was derived from the slope in the
inexion point of the curve and expressed as g of sugar
consumed per litre per day.

Dry mass

The dry cell weight of 10 mL culture samples was

determined using 0.45 lm membrane lters and a
microwave oven (180 W, 15 min) (Olsson & Nielsen,
1997).
Alcoholic beverage production
Fruit pulp

The cocoa pods were from Bahia State in the Northeast

region of Brazil. The fruits were washed and broken up
to extract the seeds and processed in the CEPEC/
CEPLAC (Cocoa Research Centre, Itabuna, BA,
Brazil). The fruits were analysed in relation to total
weight, pod:pulp and seed:pulp weight ratios. The pulp
was extracted using an automatic depulping machine
(ITAMETAL 0.5 DS, Itabuna, BA, Brazil). Samples
were taken from cocoa pulp and were characterised in
relation to total soluble solids and starch (AOAC,

 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

Elaboration of a fruit wine from cocoa D. R. Dias et al.

1990a), total sugars (Yemn & Willis, 1954), reducing

sugars (Miller, 1959), total titratable acidity and pH
(Instituto Adolfo Lutz, 1985), pectins (McCready &
McComb, 1952) and phenolics (Reicher et al., 1981).
The cocoa pulp was frozen in plastic bags and stored at
)20 C.

Chemical analysis

Analysis performed were pH, density, ethanol content,

volatile acidity, total acidity, total dry extract, free and
total SO2, determined according to the methodology
proposed by Brasil (1988) and AOAC (1990b).
Substrates and metabolites

Cocoa must

To prepare the fermenting must, cocoa pulp was

defrosted at room temperature. The cocoa pulp had an
average sugar level of 16Brix and a pH value of 3.2.
The cocoa pulp was diluted with sucrose solution to
reduce the turbidity and to adjust the sugar concentration to 22Brix. CaCO3 was added to increase the pH
value to 4.0. Pectinolytic enzyme (Ultrazym AFP-L;
Novo Nordisk Ferment Ltd, Fuglebakken, Denmark)
was also added to aid the clarication of the juice. The
enzyme concentration was used as recommended by
Novo catalogue (Novo Nordisk Ferment Ltd, 1998).
Sulphur dioxide was added: up to 200 mg of K2S2O5
per litre, to inhibit bacteria proliferation.
Fermentation conditions

Fermentation experiments were carried out in stainless

steel vats of 5 and 20 L. The fermentation temperature for cocoa wine production was approximately
22 C and no stirring was performed during any stage
of the fermentation. Inocula of the selected
S. cerevisiae were prepared as follows: one colony
from a fresh YPD plate was inoculated into 200 ml of
YPD broth and grown at 30 C until a cell density of
approximately 107 cfu mL)1 was reached. The cells
were counted and an equal amount of cells per strain
was resuspended in the same medium as used for the
fermentation. Each vat was then inoculated with
10 mL of this suspension, corresponding to a nal
cell density of 1 105 cfu mL)1, unless otherwise
stated. The maximal fermentation rate was determined
by the maximum ethanol production and decrease of
sugar content and the fermentation was considered
complete when the Brix level was stable. The CO2
production was observed during the fermentation
process. Samples were taken at the indicated points
and microbiologically and chemically analysed.
At the end of the fermentation, the vats were
transferred to a 10 C incubator to aid the sedimentation of solid material from cocoa pulp. After 10 days at
this temperature, the wine transfer was carried out with
some aeration and the beverage was incubated at 10 C
for another 30 days. After that period, another transfer
without aeration was carried out and the fruit wine was
left for another 10 days at 10 C before ltration. The
cocoa wine was then ltered through diatomaceous
earth and cellulose lters. The beverage was storage at
8 C in glass bottles fully lled to avoid oxygen
entrance.

Alcohols (ethanol and glycerol), organic acids (acetic

acid, lactic acid, malic acid, succinic acid, tartaric acid
and citric acid) and carbohydrates (glucose, sucrose and
fructose) were obtained from pulp extracts and analysed
(Schwan et al., 2001). Analyses were carried out with a
high-performance liquid chromatography system (Shimadzu, model LC-10Ai; Shimadzu Corp., Kyoto, Japan),
equipped with a dual detection system consisting of a
u.v. detector and a refractive index (RI) detector (RID
10A SPD-10Ai, Shimadzu Corp., Kyoto, Japan). A
Shimadzu ion exclusion column (Shim-pack SCR-101H,
Shimadzu Corp., Kyoto, Japan) operated at a temperature of 40 C was used to achieve chromatographic
separation. Water-soluble acids, sugars (fructose, glucose and sucrose), and ethanol were eluted with 100 mm
of perchloric acid at a ow rate of 0.8 mL min)1. The
acids were detected both by u.v. absorbance (210 nm)
and by RI, while sugars and ethanol were detected only
by RI detector. Individual sugars, acids and ethanol
were identied and their concentrations were determined by comparison with retention times and amounts
of authentic standards. All samples were examined in
triplicate. The coecient of variation was less than 5%
in each case.
Higher alcohols (1-propanol, isobutanol, 1-butanol,
isoamyl alcohol, amyl alcohol and hexanol), acetaldehyde, methanol and esters (ethyl acetate and methyl
acetate) were analysed by gas chromatography using a
Chromopack, model 511, equipped with ame ionisation detector using a capillary column of silica Carbowax 57 CB (50 m 0.22 mm, Wasw, model 52 CB,
Chrompac International, Milddelburg, the Netherlands)
(Boscolo et al., 2000). One hundred microlitres of each
sample (non-distilled) were diluted 20-fold in milli-Q
water and ltered using a nitrate-cellulose membrane
(0.20 lm pores) before injection in the gas chromotograph (GC). Operating conditions were as follows: oven
temperature was kept at 55 C for 5 min, programmed
to 100 C at 2 C min)1, kept at 100 C for 3 min,
programmed to 190  C increasing 5 C min)1, then
kept at 190 C for 30 min and then programmed to
220 C in 15 min, injector and detector temperature at
250 C, and carrier gas (N2) was at a ow rate of
1.2 mL min)1. The identication of the volatile compounds was carried out by comparing their retention
times with those of standards. One sample which
contained the internal standard and the standard
compounds at similar concentration as found in the
wine, was also treated in the same way as the wine

 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

International Journal of Food Science and Technology 2007

321

Elaboration of a fruit wine from cocoa D. R. Dias et al.

samples and the nal calculations are described on the

basis of the concentration of this reference solution.
Determination of the dierent compounds was made in
triplicate.

wines were evaluated in duplicate in each session and the

mean score of wines for each quality attribute was
computed. The results were statistically analysed using a
non-parametric test of Cherno and the Statistica
programme (Statsoft Inc., 1995).

Sensory evaluation

The nal beverage was evaluated by 50 panellists, males

and females of 1855 years of age (sta and students of
the Universities UNILAVRAS and UFLA). The panellists were selected for participation on the basis of their
preference for dry (<5 g L)1 of sugar) beverage, interest
and availability. Randomised refrigerated (10 C) samples, of 2025 mL, were served in clear, 170 mL, tulipshaped glasses; these were marked with three digit
random numbers and covered with plastic Petri dishes.
Distilled water was provided for rinsing of the palate
during the testing. Evaluations took place in the mornings between 9:00 and 10:00 a.m. and were conducted at
room temperature of 2022 C under white light.
The cocoa wines were evaluated for taste, clarity,
colour and general acceptability according to the
Hedonic scale (Moraes, 1993). This scale consists of
the comparison, punctuation, and classication of foods
and beverages of the same class or origin according to
their qualities and defects. A card of six parameters
(visual examination, smelling intensity, smelling quality,
taste intensity, taste quality and harmony) was proposed, accompanied by a scale of nine categories: dislike
extremely 1, dislike much 2; dislike moderately
3; dislike slightly 4, neither dislike nor like 5, like
slightly 6; like moderately 7; like much 8, like
extremely 9 (Moraes, 1993). The sensoric analysis was
carried out in two sensory sessions, each lasting 1 h. The

220

Results and discussion

Yeast selection

The three isolates of S. cerevisiae were selected because of

their dierences in fermentative behaviour in laboratory
scale alcoholic fermentations and they were evaluated in
relation to ability to grow and ferment under dierent
temperatures. The isolates were pregrown in YPD
medium and then cultivated in FFM medium until a
density of 107 cells mL)1 was reached. The cells were
inoculated into fresh FFM media and incubated at 18, 22
and 25 C for 64 h. All three strains showed growth when
incubated at 18 C (Fig. 1). The CA116 growth rate was
higher than that of the other two yeasts and its population
reached 1.8 108 cells mL)1 after 12 h of growth. After
this time viability decreased and a loss of viability of
about 26% was measured after 48 h in the CA116 culture.
The CA1162 isolate presented an increase in the population during the rst 12 h of incubation and then it
remained almost constant at a population of
108 cells mL)1 (Fig. 1). The CA1183 isolate showed
12 h of adaptation (lag phase) and then started to grow
at a slower rate than the other two isolates. Although the
growth was slower, the CA1183 isolate reached a
maximal population similar to CA1162, i.e. about
108 cells mL)1 (Fig. 1). The isolates CA1162 and
CA1183 had a delay in growth compared with CA116,

CA116
18 C

CA1162

200

Number of viable cells . mL1 (106)

322

CA1183
180

22 C

160
25 C

140
120
100
80
60

4 8 12 16 24 32 40 48 56 64

4 8 12 16 24 32 40 48 56 64

Time (h)

International Journal of Food Science and Technology 2007

4 8 12 16 24 32 40 48 56 64

Figure 1 Number of viable cells of three strains of Saccharomyces cerevisiae incubated at

dierent temperatures of 18, 22 and 25 C.
Cells were transferred to a fresh FFM medium every 24 h. The vertical lines indicate the
transfer to fresh fermentation medium.

 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

Elaboration of a fruit wine from cocoa D. R. Dias et al.

however, once the maximal population was reached it

remained at high values for the whole process, i.e. there
was a long stationary phase and no declining phase. No
increase in the population was observed when the
fermentation broth was renewed after 24 h, however,
there was a small increase in the number of cells after
48 h; this increase was coincident with the second renewal
of fermentation broth.
After 72 h of incubation at 18 C, the cultures were
left for 2 h at room temperature to sediment before the
supernatant was removed, and subsequently the cultures were further incubated at 22 C. The pattern of
viable cells of CA116 was similar to the one observed
when cells were incubated at 18 C. The CA116 strain
reached maximum population of 1.5 108 cells mL)1
after 16 h of incubation and then a dramatic decrease
in viability was observed over a 16 h period (Fig. 1). It
was also observed that strain CA1183 had a similar
pattern of cell density as CA116 when incubated at
22 C but with lower numbers of cells at the end of the
process (7.5 107 cells mL)1) CA1162 showed a lag
phase of 12 h and then the cell densities increased up to
1.1 108 cells mL)1 (Fig. 1). As was observed at 18
and 22 C, strain CA116 showed an increase in the
population in the rst hours of incubation at 25 C and
then a decline in cell viability (Fig. 1). CA1183 had a
small phase of adaptation and after little growth the
concentration of cells remained at 9.2 107 cells mL)1
(Fig. 1). The numbers of cells of strain CA1162 steadily
increased during the 72 h of the fermentation process
reaching 11.4 108 cells mL)1 (Fig. 1). Although
CA116 showed a rapid increase in cell numbers in the
three temperatures tested, it was also observed that this
strain showed a high decrease in cell viability as the
temperature increases (Fig. 1). The dierences in the
population of the strains CA1162 and CA1183 were
not signicant at 5% Tukey test in any of the
temperatures analysed. At 25 C, similar population
numbers of the strains CA116 and CA1162 were
observed.
The temperature of incubation may aect not only
growth and rate of fermentation but also yeast
metabolism, which will determine the chemical composition of the nal beverage. According to Jones et al.
(1981), the ideal temperature for yeast fermentation is
about 510 C above the optimum growth temperature, which is in general between 25 and 30 C.
However, there is a dierence if the nal product is
fermented or fermented and distilled. To produce
distilled beverage from sugar cane, the optimum
fermentation temperature is approximately 30 C and
the fermentation lasts 24 h (Schwan et al., 2001).
However, for wine fermentation the temperature is
usually below 25 C and the fermentation process takes
more than 10 days (Kunkee & Goswell, 1977; Vogt
et al., 1986; Torija et al., 2003).

As expected, the yeast growth and cell viability varied

according to the temperature of incubation. Our results
agree with those of Fleet & Heard (1993) who reported
that the temperature aected the development of wildtype yeast strains and suggested that various strains
showed dierences in the ability to ferment over a wide
range of temperatures.
The budding index shows that there was approximately 4565% of bud cells during all the experiments
for the three strains of S. cerevisiae analysed. These
values indicated that the fermentation conditions were
suitable for cell growth. In the CA1183 culture supernatant, after 10 min at rest, a decrease of more than
85% of the OD was detected. The CA1183 strain
showed a visible occulation after the rst change of the
culture medium.
The yeast CA1183 showed occulation behaviour
24 h after the cells were transferred to a fresh culture
medium. This strain grew and fermented as single cells,
but showed a occulation mechanism when the concentration of fermentable sugars was low. It is known that
yeast occulation is under genetic control and dominant
and recessive occulation genes and occulation suppressor genes have been described (Stratford, 1992,
1994; Teunissen & Steensma, 1995). It has been
suggested that the genes that regulate occulation might
be related to nutrient availability and metabolites
produced by yeasts during growth and fermentation
(Soares & Vroman, 2003; Soares et al., 2004; Sampermans et al., 2005). Our results suggested that the
shortage of fermentable carbon sources after 24 h
growth, combined with the presence of ethanol, which
has a positive eect on yeast occulation (Soares et al.,
2004), may be the signal compounds that induce the
onset of occulation. Flocculation favours mass sedimentation, recovery of the nal product and cells and
facilitates the batch fermentation system.
In all experiments the production of ethanol and
glycerol by the three strains of S. cerevisiae were
evaluated. There was no signicant dierence in the
ethanol production by the strains (Table 1) at the
temperatures evaluated. Glycerol is a non-volatile
compound, which has no aromatic properties, but
which signicantly contributes to wine quality by
providing sweetness and fullness (Amerine & Cruess,
1960; Eustace & Thornton, 1987; Fleet & Heard,
1993). Glycerol and succinic acid are the main
secondary products in alcoholic fermentations led by
S. cerevisiae. Glycerol amounts were approximately
10% of the total ethanol produced by strain CA1183
in all temperatures evaluated (Table 1). Glycerol
production was slightly increased with a temperature
rise from 18 to 25 C for all three strains tested,
although the strains CA116 and CA1162 showed
lower production of glycerol than strain CA1183.
No signicant dierences were observed in the values

 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

International Journal of Food Science and Technology 2007

323

324

Elaboration of a fruit wine from cocoa D. R. Dias et al.

Table 1 Results (mean SE) of chemical compounds and pH value produced by three strains of Saccharomyces cerevisiae during fermentation at
dierent temperatures in FFM medium
CA116
Temperature of
incubation
Ethanol
(g L)1)
Glycerol
(g L)1)
Acetic acid
(g L)1)
Succinic
acid (g L)1)
Acetaldehyde
(g L)1)
pH
Ethanol yield*
Time of maximum
population (h)

18 C

CA1162
22 C

25 C

18 C

CA1183
22 C

25 C

18 C

22 C

25 C

89.0 0.1155 87.0 0.1528 78.0 0.1764 92.0 0.2000 91.0 0.1453 90.5 0.2082 92.5 0.1463 92.0 0.2333 90.5 0.2423
5.53 0.0590 6.09 0.0546 7.00 0.0721 7.90 0.0797 8.10 0.0321 7.90 0.0504 9.00 0.0902 8.90 0.0643 9.00 0.0623
0.08 0.0018 0.10 0.0015 0.10 0.0012 0.12 0.0025 0.14 0.0044 0.16 0.0023 0.05 0.0009 0.07 0.0017 0.07 0.0015
0.92 0.0219 0.79 0.0153 0.71 0.0186 0.89 0.0176 0.77 0.0208 0.70 0.0145 0.95 0.0176 0.84 0.0145 0.79 0.0153
0.05 0.0011 0.05 0.0012 0.04 0.0006 0.09 0.0015 0.08 0.0009 0.09 0.0015 0.03 0.0007 0.02 0.0003 0.02 0.0002
4.1 0.0343 3.9 0.0882 3.6 0.0667 4.0 0.0284 3.8 0.0422 3.7 0.0333 4.0 0.0577 3.9 0.0326 3.7 0.0234
44.5 0.0577 43.5 0.0764 39.5 0.0882 46.0 0.1000 45.5 0.0726 45.3 0.1041 46.3 0.0726 46.0 0.1167 45.3 0.1267
12
16
16
12
16
16
24
16
12

Results are the mean of four replicates and did not vary more than 3%.
*Ethanol yield Ethanol produced (g L)1) 100/initial sugars (g L)1).

of yield of ethanol for all strains analysed (Table 1),

however the strain CA1183 showed highest ethanol
production. Several authors have reported on the
inuence of fermentation temperature on the concentrations of volatiles produced during grape fruit
fermentation and its interference in the quality of
wines (Killian & Ought, 1979; Fundira et al., 2002).
Based on the data obtained in our work strain
CA1183 was selected as inoculum for elaboration of
cocoa wine because of its high fermentation speed, fully
using up all the sugars present in the medium, its
occulation ability, its lower loss of viability, its better
yield of ethanol, its low production of volatile acids and
other fermentation by-products such as ethyl acetate
and isoamyl alcohol.
Elaboration of the beverage
Chemical characterisation of the pulp

The cocoa pulp used in this work has been removed by

mechanical processes and chemically characterised
before initiating the fermentation process. The averages
of the results in triplicate are to be found in Table 2. The
cocoa pulp had on average pH value of 3.2, which
characterised it as an acid fruit. The concentration of
total and reducing sugars was 17.4% and 10.7%
respectively (Table 2). With these amounts of sugars
the cocoa pulp could be compared with other tropical
fruits with potential for use in the beverage industry
(Morton, 1987; Franco, 1999; Alves et al., 2000). The
total soluble solid concentration in the pulp of cocoa,
determined as its Brix value, was 20.5 (Table 2). As the
determination of Brix by refractometry indicates total

International Journal of Food Science and Technology 2007

Table 2 Physico-chemical characterisation of mature cocoa pulp

Characteristics

Mean SE

Total weight (g)

543.0
Pulp (%)
26.4
Pod (%)
50.5
Seed (%)
23.1
Size- length (mm)
169.1
Diameter (mm)
94.6
Total soluble solids (Brix)
20.5
Total titratable acid (mEq L)1 in malic acid)
1.0
Brix:acidity ratio
20.5
pH
3.2
Total sugars (%)
17.4
Reducing sugars (%)
10.7
Starch (%)
0.16
Total pectin (%)
0.57
Pectin fractionated (% in relation to alcohol insoluble solids)
High methylation
4.04
Low methylation
1.49
Protopectin
1.03
Phenolics soluble in water (%)
0.17
Phenolics soluble in methanol (%)
0.15
Phenolics soluble in methanol 50% (%)
0.13

4.5826
0.2216
0.3512
0.2186
1.3077
0.8386
0.1528
0.0088
0.1880
0.0167
0.1155
0.0577
0.0012
0.0023

0.0186
0.012
0.0070
0.0012
0.0011
0.0009

soluble solids, which are not necessarily constituted of

sugars only, the nal alcohol yield might appear to be
low if based on the Brix value. The presence of
endogenous pectinases is relatively common in fruits
(Grassin & Fauquembergue, 1996), however, any
natural activity of pectinases was not detected in the
cocoa pulp. Therefore, the addition of the enzymatic
complex was necessary to reduce the viscosity of the
pulp.

 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

Elaboration of a fruit wine from cocoa D. R. Dias et al.

FRUIT PULP
22 Brix, pH 3.8
Must Preparation
Ultrazym
(0.7 mL L1)
Enzymatic treatment

Addition of sulphur
dioxide (100 mg L1)

Addition of
Bentonite (1 g L1)
106 cells mL1
Yeast inoculation

22 C, 10 days

ALCOHOLIC
FERMENTATION
10 C, 10 days
First decant

10 C, 30 days
Second decant
Cellulose plate
Filtration

FRUIT WINE

Chemical analysis
Sensory analysis

Figure 2 Flow diagram of the process of alcoholic beverage production

from cocoa pulp.

Alcoholic fermentation

The cocoa pulp was processed to get a fermentable must

from which the alcoholic beverage was produced. The
owchart represented in Fig. 2 shows the dierent stages
to which the must of the fruit was submitted up till the
nal product. Before the beginning of the fermentative
process, sucrose, pectinolytic enzymes, sulphite and
bentonite had been added to the cocoa pulp. To attain
22Brix, 10 L of sucrose solution (350 g L)1) had to be
added to an equal volume of 10 L of cocoa pulp. The
addition of the enzymatic complex Ultrazym AFP-L
in the concentration of 0.7 ml L)1 favoured the process
of clarication of the wine of cocoa. Gomez-Ruiz et al.
(1988) had obtained similar results in the clarication of
apple juices treated with endopolygalacturonases. To
limit the bacterial growth 100 mg of SO2 was added per
litre of must (about 200 mg of K2S2O5). These values,
although considered high to initiate the fermentation,
had been adopted because of low pH value of the cocoa
must (Hashizume, 2001). During the counting of the

population of viable cells, in the period of fermentation,

the presence of bacteria was not observed, indicating
that the use of metabisulphite was ecient in the control
of the bacterial growth during the fermentative process
(data not shown). Another evidence of absence of
contamination, in this case for lactic acid bacteria was
that lactic was not detected, whereas malic acid was
found in the nal beverage of cocoa (Table 3). However,
as a result of the SO2 concentration used for the
elaboration of the cocoa wine a delay in the beginning of
the fermentation of about 12 h was observed. These
observations agree to those described by Gerbaux &
Meurgues (1995). These authors had a delay in the
beginning of the fermentation when they treated the
grape must with concentrations of SO2 higher than
80 mg L)1. It was also observed that once the fermentation had started, its course was not aected by
sulphite, as was also described by Gerbaux & Meurgues
(1995).
To facilitate the non-fermentable solids to sediment,
1% of bentonite was added to the must. The bentonite
was suspended in water in a 10% concentration to
facilitate the dispersion of the clay in the must. The main
eect of the bentonite was the precipitation of proteinaceous material by adsorption and neutralisation,
involving denaturation of oxidative enzymes (Manfredini, 1989). The bentonite introduced to the cocoa must
in the pre-fermentative phase provided a better clarication of the beverage by facilitating the sedimentation
of the solids of the must, thereby improving decanting
and ltration in later stages.
The fermentation temperature selected in the previous
phase was 22 C and strain CA1183 was chosen as
inoculum for the fermentation process. The inoculum of
S. cerevisiae CA1183 of 106 cells mL)1 showed to be
ecient as the ethanol concentration after 10 days of
fermentation was close to 9.6% w/v (12% v/v). These
results could be compared with the ones reported by
Oliveira et al. (2001), who found a value of 7.5% w/v
when fermenting pulp of pupunha with an initial
24Brix. The fermentation process took 50 days and
after this period the beverage was ltered and stored in
glass bottles at 4 C.
Chemical analyses of the beverage

The chemical analyses of the nal beverage produced

from cocoa pulp showed that the beverage produced is
similar to dry table wines (Table 4). Although cocoa and
grape are dierent fruits, the steps to produce the
alcoholic beverage were very similar. All the steps
required good hygiene conditions and the results showed
that they were satisfactory in relation to the purposed
methodology.
Sucrose, glucose or fructose were not detected in the
nal beverage of cocoa wine (Table 3), which charac-

 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

International Journal of Food Science and Technology 2007

325

326

Elaboration of a fruit wine from cocoa D. R. Dias et al.

Table 3 Concentration (mean SE) of organic compounds determined with HPLC and GC (acids, sugars and alcohols) in the
fermented beverage from cocoa pulp
Compounds

Concentration in cocoa wine

)1

HPLC analyses (g L )
Carbohydrates
Sucrose
Glucose
Fructose
Alcohols
Ethanol
Glycerol
Acids
Acetic
Citric
Lactic
Malic
Oxalic
Succinic
Tartaric
GC analyses (mg L)1)
Aldehydes
Acetaldehyde
Esters
Methyl acetate
Ethyl acetate
Alcohols
Methanol
Higher alcohols
Propanol
Isobutanol
Butanol
Isoamyl alcohol
Amylic alcohol
Hexanol
Total of higher alcohols

12.3 0.0110
4.6 0.0068
0.0058
0.0306
0.0137

Indicators

Maximum Minimum
10.0
()
55.0
()
()
()
()
()
20.1
()

Cocoa alcoholic
beverage
(mean SE)
12.5
n.d.
98.5
6.5
n.d.
n.d.
0.06

0.0667
0.8192
0.0500

0.0006

n.d.
()
3.6 0.0173

n.d., not detected; () not established.

*Source: Brasil (1988).

0.0120
0.0067

n.d.
n.d.
350.7 0.1342
n.d.
44.8
90.4
n.d.
498.8
n.d.
n.d.
634

Limits*

Ethyl alcohol % (v v)
13.0
0.35
Methyl alcohol (g L)1)
Total acid (mEq L)1)
130.0
20.0
Volatile acids (mEq L)1)
1.0
Total sulphate, as K2SO4 (g L)1)
0.20
Total chloride, as NaCl (g L)1)
Total SO2 (g L)1)
0.35
Total sugars
5.0
Dry wine (g L)1)
()
Sweet wine (g L)1)
pH
()

n.d.
n.d.
n.d.

1.1
5.5
n.d.
1.4
n.d.
2.0
0.7

Table 4 Analytical values found in alcoholic beverage from cocoa

0.0338
0.0153
0.1337

0.1193

n.d., not detected.

terised it as a dry beverage (Brasil, 1988). Methanol was

not detected either in the samples of must or in the nal
product. Despite the high concentration of SO2 in the
beginning of the fermentation of the cocoa must
(100 mg L)1), it could be observed it was eliminated
during the fermentative process and decanting, not
leaving residual concentrations, which could disqualify
the avour of the nal beverage.
Tables 3 and 4 present analyses of compounds that
are important with respect to the avour and the aroma
of fermented beverages. The results show that the total
amount of higher alcohols for the cocoa beverage was
about 0.6 g L)1 (Tables 3 and 4). This level was higher
than that reported by Vogt et al. (1986), who showed a
variation between 0.1 and 0.3 g L)1 of higher alcohols in
wines. The amount of these alcohols in wines is
inuenced by the presence or absence of SO2, as
reported by Herraiz et al. (1989). Although these
authors described a positive correlation between the

International Journal of Food Science and Technology 2007

presence of the SO2 during the fermentation and a high

acetaldehyde concentration in the wine, this did not
occur in the cocoa beverage, where the presence of
acetaldehyde was not detected (Table 3). The absence of
the acetaldehyde could have been due to the low pH
(3.8) value of the must, as also observed by Cleto &
Mutton (2004). The values of ethyl acetate found in this
work (Table 3) are higher than those described by Berry
& Watson (1987) and are similar to values found by
Cleto & Mutton (2004) in fermented beverage from
sugar cane, orange and grape. The glycerol concentration, about 4.6 g L)1, was close to the values of 6 and
10 g L)1, suggested by Vogt et al. (1986) as characteristic to confer body and texture to the beverage. In
Table 3 there are some remarkably high values of
isoamyl alcohol (about 499 mg L)1) and ethyl acetate
(about 351 mg L)1) compounds, which contribute to the
fruity avour in the nal beverage.
Sensory analysis of the beverage

After the chemical analyses, the beverage was submitted

to sensory analysis to assess its acceptance among the
consumers. Table 5 presents notes attributed to
the beverage by 50 untrained tasters, designated in the
hedonic scale of nine points. Similar averages were
recorded for the four evaluated attributes whereas the
aroma is the one with slightly higher value, followed by
general appearance, taste and overall aspects, with
respective notes of 7.7, 7.6, 7.5 and 7.5 (Table 5). The
aroma was also responsible for the biggest maximum
note frequency, getting 16 votes (35.5% of the tasters)
(Table 5). The attribute of overall aspects, which
corresponds to the global acceptance of the beverage
by part of the tasters, reached a higher frequency (19
tasters, that is, 42.2% of the total) in note 8, what meant

 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

Elaboration of a fruit wine from cocoa D. R. Dias et al.

Table 5 Frequency and average notes for the attributes of sensory

analysis
Frequency and average notes
Attribute

Mean SE

Appearance
Aroma
Taste
Overall

3
1
1

5
1
5
4

14
14
9
11

12
11
17
19

12
16
11
8

7.6
7.7
7.5
7.5

0.17
0.20
0.24
0.17

1 dislike extremely; 9 like extremely.

like very much the beverage (Table 5). The taste of the
beverage reached the lesser average (7.5) what reected
that the tasters had moderately liked the beverage.
According to Rapp & Mandery (1986), the trend is that
the beverage loses sensorial qualities when the total
concentration of higher alcohols is over 400 mg L)1. In
the evaluation of the cocoa beverage, according to the
methodology used, the 633.0 mg L)1 of high alcohols
had not brought about a negative inuence on aroma, as
aroma was the most appreciated attribute.
The expression of acceptance and non-acceptance of
the beverage is presented in Table 6, in function of the
grades attributed by the tasters. It can be observed
that in relation to the appearance, none of the tasters
was displeased with the beverage and the percentage
of acceptance of this attribute (they had at least liked
very much) was of 53%. For the attributes aroma
and avour, about 7% of the tasters were displeased
with the beverage, while, also for both, about 60%
liked it much. Table 6 also shows that, in relation to
the general aspects, 60% of the tasters had liked the
beverage much, while 2% was displeased. A nonparametric test known as faces of Cherno was
carried out to establish the approval of the beverage
with the tasters (Fig. 3). This test made clearer the
dierences reached among the attributes of the beverage by using a visual recognition. The values
attributed to the cocoa beverage had been compared
with a standard scale normal. From the presented
diagrams a homogeneity in conferred grades to each
one of the attributes and in relation to the global

Table 6 Percentage of acceptance and rejection of the fermented

beverage from cocoa

Cocoa

Normal

Figure 3 Representation of the acceptability of cocoa fermented

beverage through Chernoffs face where size of the face is overall
rating, width of the nose is aroma, curvature of the mouth is taste and
size of the eyes is appearance.

acceptance of the beverage was observed (overall

aspects).
Conclusions

From the data reported in this experiment, comparing

three strains of S. cerevisiae (CA116, CA1162 and
CA1183), it could be concluded S. cerevisiae strain
CA1183, selected in this work, appeared to be capable to
ferment all the sugars of the cocoa must and it also
possessed the occulation characteristic. The processes
normally used in fermented beverage manufacture could
be adapted for the elaboration of the fermented cocoa
pulp. The dilution of cocoa pulp with sucrose solution,
to improve the uidity of must, diminished the concentration of marker compounds of the beverages aroma,
not intervening with the production of ethanol, as the
formation of ethanol follows a dierent pathway. The
use of the enzymatic complex and the solution of
bentonite during all the fermentative process brought
satisfactory results for the clarication of the beverage,
as well as the SO2, that in the concentration applied
inhibited the growth of bacteria and did not interfere
with the quality of the nal beverage. The sensory
analysis revealed a good acceptance, mainly when taking
into account the fact that the tasters were not familiar
with the cocoa-derived beverage. The methodology
described here is simple, and can be implanted in a
relatively small space and without high cost. It was
observed that, from the acceptability of the beverage,
this technology can be one of the alternatives for the use
of the cocoa pulp, or a new industrial outlet for this
fruit.
Acknowledgments

Attribute

14 (rejection)
(% of tasters)

89 (acceptance)
(% of tasters)

Appearance
Aroma
Taste
Overall

0.0
6.7
6.7
2.2

53.3
60.0
62.2
60.0

The work was nanced by European Union (ERB IC18

CT97 0182). DRD thanks to CNPq for the scholarship
during his masters course and the UNILAVRAS for
their support during this work. The authors are grateful
to Prof. Frank Rombouts for his help with the editing of
this manuscript.

 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

International Journal of Food Science and Technology 2007

327

328

Elaboration of a fruit wine from cocoa D. R. Dias et al.

References
Akubor, P.I. (1996). The suitability of African bush mango juice for
wine production. Plant Foods and Human Nutrition, 49, 213219.
Alves, R.E., Filgueiras, H.A.C. & Moura, C.F.H. (2000). In:
Caracterizacao de frutas nativas da America Latina (edited by R.E.
Alves, H.A.C. Filgueiras & C.F.H. Moura). Vol. 9 (Serie Frutas
Nativas). Pp. 1518. Brazil: Jaboticabal.
Amerine, M.A. & Cruess, W.V. (1960). The Technology of Winemaking. Pp. 213302. Westport, CT: Avi.
Association of Ocial Analytical Chemists (1990a). Ofcial Methods
of Analyses of AOAC. 15th edn, Vol. 2, Chapter 37. Pp. 915922.
Washington, DC: AOAC.
Association of Ocial Analytical Chemists (1990b). Ofcial Methods
of Analyses of AOAC. 15th edn, Vol. 2, Chapter 28. Pp. 744.
Washington, DC: AOAC.
Ayogu, T.E. (1999). Evaluation of the performance of yeast isolate
from Nigerian palm wine in wine production from pineapple fruit.
Bioresource Technology, 69, 189190.
Berry, D.R. & Watson, D.C. (1987). Production of organoleptic
compounds. In: Yeast Biotechnology (edited by D.R. Berry, I. Russell
& G.G. Stewart). London: Allen & Unwin.
Boscolo, M., Bezerra, C.W.B., Cardoso, D.R., Neto, B.S.L. & Franco,
D.W. (2000). Identication and dosage by HRGC of minor alcohols
and esters in Brazilian sugar cane spirit. Journal of Brazilian
Chemical Society, 11, 8690.
Brasil (1988). Leis, Decretos, etc. Lei n. 7678, de 08 de out. de 1988.
Bras lia: Ministerio da Agricultura e do Abastecimento, [Internet
document] URL http://oc4j.agricultura.gov.br/agrolegis/do/consultaLei?opviewTextual&codigo189. Accessed 7/08/2005.
Cappello, M.S., Bleve, G., Grieco, F., Dellaglio, F. & Zacheo, G.
(2004). Characterization of Saccharomyces cerevisiae strains isolated
from must of grape grown in experimental vineyard. Journal of
Applied Microbiology, 97, 12741283.
Cleto, F.G.V. & Mutton, M.J.R. (2004). Rendimento e composicao
das aguardentes de cana, laranja e uva com utilizacao de lecitina
no processo fermentativo. Ciencia e Agrotecnologia, 28, 577
584.
Colagrande, O., Silva, A. & Fumi, M.D. (1994). Recent applications of
biotechnology in wine production. Biotechnology Progress, 10, 218.
Demain, A.L. (2000). Microbial biotechnology. Trends in Biotechnology, 18, 2631.
Dias, D.R., Schwan, R.F. & Lima, L.C.O. (2003). Metodologia para
elaboracao de fermentado de caja (Spondias mombin L). Ciencia e
Tecnologia de Alimentos, 23, 342350.
Eustace, R. & Thornton, R.J. (1987). Selective hybridization of wine
yeast for higher yields of glycerol. Canadian Journal of Microbiology,
33, 112117.
FAOSTAT Data (2005). FAOSTAT Database Collection: Agricultural
Data. [Internet document] URL http://faostat.fao.org/faostat/
servlet/XteServlet3?Areas%3E862&Items661&Elements51&
Years2004&FormatTable&XaxisYears&YaxisCountries&
Aggregateover_items&Calculate&DomainSUA&ItemTypes
Production.Crops.Primary&languageEN. Accessed 7/08/2005.
Fleet, G.H. & Heard, G.M. (1993). Yeast-growth during fermentation.
In: Wine Microbiology and Biotechnology (edited by G.H. Fleet). Pp.
2754. Chur: Harwood Academic Publishers.
Franco, G. (1999). Tabela de composicao qumica dos alimentos. 9th
edn. Pp. 307. Sao Paulo: Atheneu.
Freire, E.S., Schwan, R.F. & Mororo, R.C. (1999). The cocoa pulp
agroindustry and the use of its residues in Bahia: progress achieved
in the last ten years. In: Proceedings of the 12th International Cocoa
Research Conference. Pp. 10131020. Kuala Lumpur: Cocoa
Producers Alliance.
Fundira, M., Blom, M., Pretorius, I.S. & van Rensburg, P. (2002).
Selection of yeast starter culture strains for the production of marula
fruit wines and distillates. Journal of Agriculture and Food
Chemistry, 50, 15351542.

International Journal of Food Science and Technology 2007

Garcia, E.H., Cahanap, A.C. & Cabrera, M.P. (1976). Organoleptic

and chemical properties of four new Philippine fruit wines II
(banana, buko water, pineapple and tomato). The Philippine Journal
of Plant Industry, 40, 2935.
Gerbaux, V. & Meurgues, O. (1995). Inuence du sultage et du
debourbage des mouts sur lelaboration et la qualite des vins de
chardonnay. Revue des nologues, 78, 1518.
Gomez-Ruiz, L., Garcia-Garibay, M. & Barzana, E. (1988). Utilization of endo-polygalacturonase from Kluyveromyces fragilis in the
clarication of apple juice. Journal of Food Science, 53, 12361240.
Grassin, C. & Fauquembergue, P. (1996). Fruit juices. In: Industrial
Enzymology, 2nd edn (edited by T. Godfrey & S. West). Pp. 225
264. Hampshire, UK: Macmillan Press.
Hansen, J. & Kielland-Brandit, M.C. (1996). Modication of
biochemical pathways in industrial yeasts. Journal of Biochemistry,
49, 112.
Hashizume, T. (2001). Tecnologia do vinho. In: Biotecnologia Industrial: Biotecnologia na producao de alimentos (edited by E. Aquarone, W. Borzani, W. Schmidell & U.A. de Lima). Vol. 4. Pp. 2168.
Sao Paulo: Edgard Blucher.
Heard, G.M. & Fleet, G.H. (1988). The eects of temperature and pH
on the growth of yeast species during the fermentation of grape
juice. Journal of Applied Bacteriology, 65, 2328.
Herraiz, T., Martin-Alvarez, P.J., Reglero, G., Herraiz, M. &
Cabezudo, M.D. (1989). Dierences between wines fermented with
and without sulphur dioxide using various selected yeasts. Journal of
the Science of Food Agriculture, 49, 249258.
Instituto Adolfo Lutz (1985). Normas analticas do Instituto Adolfo
Lutz: Metodos qumicos e fsicos para analises de alimentos, 3rd edn,
Vol. 1. Pp. 27. Sao Paulo: Instituto Adolfo Lutz.
Jones, R.P., Pamment, N. & Greeneld, P.F. (1981). Alcohol
fermentation by yeasts: the eect of environmental and other
variables. Process Biochemistry, 16, 4249.
Killian, E. & Ought, C.S. (1979). Fermentation ester formation and
retention as aected by fermentation temperature. American Journal
of Enology and Viticulture, 30, 301305.
Kunkee, R.E. & Goswell, R.W. (1977). Table wines. In: Alcoholic
Beverages (edited by A.H. Rose). Vol. 1. Pp. 315. London:
Academic Press.
Lord, P.G. & Wheals, A.E. (1980). Asymmetrical division of
Saccharomyces cerevisiae. Journal of Bacteriology, 142, 808818.
Manfredini, M. (1989). Coadiuvanti enologici: bentonite. Vignevini, 4,
4346.
McCready, R.M. & McComb, E.A. (1952). Extraction and determination total pectin material in fruits. Analytical Chemistry, 24, 1586
1588.
Miller, G.L. (1959). Use of dinitrosalicylic acid for determination of
reducing sugar. Analytical Chemistry, 31, 426428.
Moraes, M.A.C. (1993). Metodos para avaliacao sensorial de alimentos,
8th edn. Pp. 93. Campinas, Brazil: Unicamp.
Morton, J. (1987). Yellow mombin. In: Fruits of Warm Climates
(edited by J.F. Morton). Pp. 245248. [Internet document] URL
http://www.hort.purdue.edu/newcrop/morton/yellow_mombin_ars.
html. Accessed 4/03/2005.
Novo Nordisk Ferment Ltd (1998). Ultrazym AFP-L: a Special
Enzyme Preparation for Liquefaction of Fruit. Pp. 3. West Sussex:
Novo Nordisk Ferment Ltd (Manufacturer catalog).
Oliveira, L.P., Maeda, R., de Andrade, J.S., Pereira, N., Jr, da
Carvalho, S.M.S. & Astol Filho, S. (2001). Processo fermentativo
para producao de bebida alcoolica de pupunha. Biotecnologia,
Ciencia e Desenvolvimento, 19, 5054.
Olsson, L. & Nielsen, J. (1997). On-line and in situ monitoring of
biomass in submerged cultivations. Trends in Biotechnology, 15,
517522.
Ostergaard, S., Olsson, L. & Nielsen, J. (2000). Metabolic engineering
of Saccharomyces cerevisiae. Microbiology and Molecular Biology
Reviews, 64, 3450.

 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

Elaboration of a fruit wine from cocoa D. R. Dias et al.

Rapp, A. & Mandery, H. (1986). Wine aroma. Experientia, 42, 873

884.
Reicher, F., Sierakowski, M.R. & Correa, J.C.B. (1981). Determinacao
espectrofotometrica de taninos pelo reativo fosfotungstico fosfomol bdico. Arquivos de Biologia e Tecnologia, 24, 401411.
Rose, A.H. (1977). History and scientic basis of alcoholic beverage
production. In: Alcoholic Beverages (edited by A.H. Rose). Vol. 1.
Pp. 137. London: Academic Press.
Sampermans, S., Mortier, J. & Soares, E.V. (2005). Flocculation onset
in Saccharomyces cerevisiae: the role of nutrients. Journal of Applied
Microbiology, 98, 525531.
Schwan, R.F. (1988). Cocoa fermentations conducted with a dened
microbial cocktail inoculum. Applied and Environmental Microbiology, 64, 14771483.
Schwan, R.F. & Wheals, A.E. (2004). The microbiology of cocoa
fermentation and its role in chocolate quality. Critical Reviews in
Food Science and Nutrition, 44, 205222.
Schwan, R.F., Mendonca, A.T., Silva, J.J., Jr, Rodrigues, V. &
Wheals, A.E. (2001). Microbiology and physiology of cachaca
(aguardente) fermentations. Antonie van Leeuwenhoek, 79, 8996.
Selli, S., Cabaroglu, T. & Canbas, A. (2003). Flavour components of
orange wine made from a Turkish cv. Kozan. International Journal
of Food Science and Technology, 38, 587592.
Soares, E.V. & Vroman, A. (2003). Eect of dierent starvation
conditions on the occulation of Saccharomyces cerevisiae. Journal
of Applied Microbiology, 95, 325330.
Soares, E.V., Vroman, A., Mortier, J., Rijsbrack, K. & Mota, M.
(2004). Carbohydrate carbon sources induce loss of occulation of
an ale-brewing yeast strain. Journal of Applied Microbiology, 96,
11171123.

Statsoft Inc. (1995). Statistica for Windows [Computer Program

Manual]. Tulsa, OK: Statsoft Inc.
Stratford, M. (1992). Yeast occulation: a new perspective. Advances
in Microbiology and Physiology, 33, 172.
Stratford, M. (1994). Genetic aspects of yeast occulation: in
particular, the role of FLO genes in the occulation of
Saccharomyces cerevisiae. Colloids and Surfaces B: Biointerfaces,
2, 151158.
Teunissen, A.W.R.H. & Steensma, H.Y. (1995). Review: the dominant
occulation genes of Saccharomyces cerevisiae constitute a new
subtelomeric gene family. Yeast, 11, 10011013.
Torija, M.A., Roze`s, N., Poblet, M., Guillamonm, J.M. & Mas, A.
(2003). Eects of fermentation temperature on the strain population
of Saccharomyces cerevisiae. International Journal of Food Microbiology, 80, 4753.
Ubeda-Iranzo, J.F., Gonzalez-Magana, F. & Gonzalez-Vinas, M.A.
(2000). Evaluation of the formation of volatiles and sensory
characteristics in the industrial production of white wines using
dierent commercial strains of the genus Saccharomyces. Food
Control, 11, 143147.
Vidya, B. & Neela, B. (2004). Eects of dierent pectolase concentration and yeast strains on carambola wine quality in Trinidad, West
Indies. Fruits, 59, 131140.
Vogt, E., Jakob, L. & Lemperle, E. (1986). El vino: obtencion,
elaboracion y analisis. 2nd edn. Pp. 294. Zaragoza, Spain: Acribia.
Walker, G.M. (1998). Yeast Physiology and Biotechnology. Pp. 350.
Chichester, UK: Wiley.
Yemn, E.W. & Willis, A.J. (1954). The estimation of carbohydrates in
plant extracts by anthrone. The Biochemical Journal, 57, 508514.

 2007 The Authors. Journal compilation  2007 Institute of Food Science and Technology Trust Fund

International Journal of Food Science and Technology 2007

329