Bioresource Technology 92 (2004) 71–78

Mango seed uses: thermal behaviour of mango seed almond fat

and its mixtures with cocoa butter
a,* b,1
J.A. Sol
ıs-Fuentes , M.C. Dur

an-de-Baz

ua
a
Instituto de Ciencias B

asicas, Universidad Veracruzana, Av. Dos Vistas s/n carretera Xalapa-Las Trancas,
91000 Xalapa, Ver., Mexico
b
Departamentos de Alimentos y Biotecnolog
ıa y de Ingenier
ıa Qu
ımica, PIQAyQA, Facultad de Qu
ımica,
UNAM, Edif. ‘‘E’’ planta baja, 04510 Mexico, DF
Received 27 April 2003; accepted 10 July 2003

Abstract
This paper deals with the physicochemical characterization, including thermal behaviour, by differential scanning calorimetry of
mango seed almond fat (MAF), alone and in mixtures with cocoa butter (CB). Results showed that mango almond seeds contain
about 5.28–11.26% (dw) of fat. The refraction index is 1.466, the saponification index 189.0 and the iodine index 41.76. Fatty acids
found in MAF are oleic, stearic, and palmitic acids (40.81%, 39.07% and 9.29% (w/w), respectively) as well as smaller amounts of
linoleic, with arachidic, behenic, lignoceric, and linolenic acids, among others. Calorimetric analysis showed that MAF crystallizes
between 14.6 and )24.27 °C with a DHc of 56.06 J/g and melts between )17.1 and 53.8 °C, with fusion maxima at 18.54 °C and
40.0 °C for the a and b polymorphic forms. Their fusion enthalpies are 70.12 and 115.7 J/g. The MAF solids content profile is very
similar to that of CB, both in stabilized and non-stabilized samples. The mixing compatibility was analyzed using isosolids curves of
mixtures of different compositions.
Ó 2003 Elsevier Ltd. All rights reserved.

Keywords: Mango almond fat; Cocoa butter; Fat blends; Phase change; Fusion enthalpies; Isosolids diagrams

1. Introduction Developing countries, the traditional producers of

The identification and application of new materials is
important for the development of new technological
approaches towards the use of traditional raw materials.
Agroindustrial residues may become unconventional
sources for these traditional raw materials (Sol
ıs-Fuen-
tes, 1998).
Vegetable fats and oils are widely used in the food,
pharmaceutical, cosmetic and chemical industries and
are normally obtained from oilseeds such as sesame
seed, soy bean, cotton seed and oil (Gutcho, 1979; Bir-
ker and Padley, 1987; O’Brien, 1998). Among these fat
and oil sources, cocoa butter (CB) is highly appreciated
because of its physical and chemical characteristics.
*
Corresponding author. Tel.: +52-2288-12-5745; fax: +52-2288-12-
9963.
E-mail addresses: jsolis@uv.mx (J.A. Sol
ıs-Fuentes), mcduran@
servidor.unam.mx (M.C. Dur
an-de-Baz
ua).
1
Fax: +52-55-5622-5303.
these seeds as well as of other agricultural products, are
currently confronting environmental challenges due to
the excess of agricultural wastes associated with pro-
duction and exportation trends. To help reverse these
problems, attention has been drawn towards the possi-
ble use of some of these agroindustrial wastes as non-
conventional sources of useful materials, including the
industrial use of fats and oils (Ben-Gera and Kramer,
1969; Maurya and Chawdhary, 1977; El-Zanati and
Zaher, 1990).
The physical, chemical and nutritional properties of
fats and oils are limiting factors for their use in certain
industrial sectors. Thus, obtaining nutritionally sound
products for the food industry will greatly depend on the
physical and/or chemical characteristics of the fat and
oil formulations. Among these physical and chemical
characteristics, thermal behaviour and phase changes
are particularly important. In the food industry, for
example, vegetable oil hydrogenation modifies its ther-
mal behaviour and stability, making possible its use as a

0960-8524/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2003.07.003
72 J.A. Sol
ıs-Fuentes, M.C. Dur
an-de-Baz
ua / Bioresource Technology 92 (2004) 71–78
substitute for traditional saturated fats of animal origin
as lard, butter, etc. (Bertoli et al., 1995; O’Brien, 1998).
However, new trends are starting to use less processed
fats and oils because of the findings on the role played by
trans fatty acids in some human health aspects (Khosla
and Sundram, 1996; Sol
ıs-Fuentes and Dur
an-de-Baz
ua,
2001). These trends open up new perspectives for the use
of natural-origin vegetable fat mixtures as alternative
sources of fatty products with adequate physical char-
acteristics, such as melting or fusion point, solid con-
tents, and consistency. These mixtures should also
possess adequate nutritional properties (Williams et al.,
1997; Kloek et al., 2000; Toro-V
azquez et al., 2000).
Mango seed almonds fat (MAF) has been the subject
of several research studies because some of its char-
acteristics show resemblance to those of CB (Nara-
shima-Char et al., 1977; Lakshminarayana et al., 1983;
Dhinigra and Kapoor, 1985; MdAl
ı et al., 1985; MdAli
and Dimick, 1994; Jim
enez-Berm
udez et al., 1995; Sol
ıs-
Fuentes, 1998). However, there are no studies of thermal
and phase behaviour of MAF from its different varieties,
and particularly the physicochemical properties of MAF
from Manila variety are scarce. The Manila mango
(Mangifera indica, L.) variety is a widely cultivated and
processed agricultural product in Mexico, due to its
particularly delicious and delicate flavour, its lack of
fibres, and the smallness of its seed. Considering that
Manila mango might be an interesting source of fat, the
aim of this research was: to determine the general
composition of mango, var. Manila, its almond seed
characteristics, including its fat (MAF) composition and
some specific properties. The MAF thermal and phase
behaviour were studied and its interactions in mixtures
with cocoa butter (CB) were investigated using differ-
ential scanning calorimetry (DSC) and drawing the re-
sulting isosolids diagrams.
2. Methods
2.1. Sampling
Mangoes from Mangifera indica, L., Manila variety
were obtained from a plantation located in the coastal
mango-producing zone of the State of Veracruz in
Mexico. Two stages of physiological ripeness or matu-
rity were selected, pre-mature and mature, in order to
compare the fat contents and the physical and chemical
characteristics of both stages.
2.2. Fruits characterization
The size and proportions of constitutive parts as well
as the chemical composition for each stage of ripeness
were determined in randomly sampled fruits. Three ex-
emplars for each maturation stage were taken, their
parts removed (skin, pulp and seed and, from seeds, the
almonds) and weighed. Pulp chemical analyses con-
firmed the ripeness stage. Moisture content (Horwitzs,
1980), ash content (Horwitzs, 1980), titrable acidity (by
neutralization), reducing and total sugars content (Lane
and Eynon method, cited by Southgate, 1991), pH value
(potentiometer measurement), and tannins content
(Folin–Denis method, cited by Horwitzs, 1980) analyses
were performed. Almonds were also characterized by
measuring their content of moisture, ash, fat, crude
fibre, protein (Horwitzs, 1980), tannins (Folhin–Denis
method, cited by Horwitzs, 1980), and cyanides (alkaline
titration).
2.3. Crude fat extraction and further purification
Once free of their external protective layer, seed al-
monds of mature fruits were dehydrated in an oven at
60 °C until they reached a moisture content of about
10%. Once dried, they were manually crushed and
ground and stored in polyethylene bags at 5 °C until use.
Fat extraction from dried almond seed meal was per-
formed in a Soxhlet apparatus utilizing hexane as sol-
vent. A 6-h extraction was carried out for total removal
of fat. Solvent was removed from the resulting product
under vacuum using a rotavapour. Fat was kept away
from light and air at 5 °C until processing and analysis
took place.
2.4. Fat purification
MAF of mature fruits was purified using an adapta-
tion of the Wesson method (Mehlenbacher, 1970): crude
fat was dissolved in petroleum ether in a ratio 1:5 (w/v).
Potassium hydroxide at 14% was added to dissolved fat
(1 ml KOH per gram of dissolved fat) with vigorous
stirring for 3 min. Then 50% ethyl alcohol was added in
a ratio of 1:2.5 (w/v) while stirring and it was left to
stand until phase separation. The ether phase contained
the neutral fat. The soap alcohol phase was repeatedly
reextracted with ether. Ether extracts were processed in
a rotavapour to recover purified fat. Purified fat was
kept in a freezer (5 °C) away from light until processing
and/or analysis.
2.5. Physicochemical properties
MAF refraction index, iodine index, saponification
index, etc. were analyzed using officially approved
techniques (Horwitzs, 1980; Hendrikse et al., 1994).
2.6. Fatty acids pattern
Fatty acids present in the MAF were analyzed using
gas chromatography coupled to mass spectroscopy,
adapting the Bannon et al. (1982) and Hendrikse et al.
 J.A. Sol
ıs-Fuentes, M.C. Dur
an-de-Baz
ua / Bioresource Technology 92 (2004) 71–78
(1994) methodology. Fat was converted into the corre-
sponding methyl esters of its fatty acids. These were
dissolved in heptane, separated by packed column
chromatography and passed to a mass spectrometer for
identification through comparison of the spectra with
previously injected standards.
2.7. Fatty acids methylic esters preparation
The boron trifluoride method (AOAC, 1995) was
used for fatty acid methyl ester preparation and ap-
proximately 1 ml of the resulting heptane layer was
transferred to a glass tube and injected into the chro-
matograph. Chromatograms and mass spectra for each
fatty acid methyl ester were obtained.
2.8. Gas chromatography–mass spectroscopy
Fatty acid methyl esters analysis was performed in a
Hewlett Packard gas chromatograph model 5890 series
II, coupled to a mass spectrometer with an electronic
ionization detector. The system had an AT Silar capil-
lary column 30 m long, 0.32 mm internal diameter and
0.25 lm film thickness. Compound identification was
carried out using a data base complemented with in-
ternal standards of fatty acids methyl esters. Helium was
the carrier gas with a flow of 0.4 mL/min and a pressure
at 50 °C of 38 kPa. Injector temperature was 250 °C and
starting oven temperature was 100 °C. The temperature
programme was 2 min at 100 °C increased by 4 °C/min
to 180 °C which was maintained for a final time of 5
min. The mass analyzer range was 10:450 m/z.
2.9. Cocoa butter
Commercially available pure CB was characterized in
the same way as MAF (iodine index and fatty acids
analysis).
2.10. Thermal profile
Samples of purified MAF, and purified CB, were fed
to a differential scan calorimeter (Dupont Inst. DSC
910) equipped with a thermal analysis data station
(Thermal Analysis 2100). The purging gas used was
nitrogen at a flow of 20 mL/min. The instrument was
calibrated with indium (melting point, 156.6 °C; DHf ,
28.45 J/g). Samples between 5 and 15 mg were weighed
in a thermobalance (TA), in aluminium SFI capsules
with a precision of ±0.1 mg, which were then hermet-
ically sealed. An empty sealed capsule was used as
reference. In order to find the thermal history of the
samples, many MAF and CB samples were previously
tempered by heating at 90 °C for 5 min and then cooled
at 24 °C for 24 h. These samples were then stored for
15 days at 5 °C.
73
The temperature programme for the calorimeter was:
(a) Cooling to )60 °C and heating at 10 °C/min to 90 °C.
Registration of the fusion profile.
(b) Heating at 90 °C for 10 min, and cooling at 10 °C/
min to )60 °C. Registration of the crystallization
profile.
(c) Heating at 90 °C, cooling at )60 °C for 10 min, and
heating at 10 °C/min to 90 °C. Registration of the
profile and enthalpy of fusion.
The enthalpies of fusion needed to melt the fat sample
crystals, as well as the crystallization enthalpies needed
to solidify them, were calculated using the area below
the integration curve and the thermogram baselines.
Temperature ranges of fusion and crystallization were
determined using the thermograms, through the identi-
fication in the curves of the onset and offset temperatures
for the phase changes.
2.11. Contents of solid fat
The amounts of solids in the fats as a function of
the temperature were calculated using DSC experiments,
the Lambelet and Raemy (1983), methodology and the
computational package Origin (Microcal, 1999). Since
solid fat contents (SFC) depend on temperature these
values may be presented as a function of temperature.
Partial areas of the thermograms were calculated and
correlated with solids percentages considering that at
)60 °C, samples are 100% solids.
2.12. Thermal behaviour of mixtures MAF–CB
A matrix was created with mixtures of MAF and CB
in different proportions. Each sample was prepared with
its duplicate and fusion profiles were studied using DSC,
following the methodology and equipment described
above. Samples were tested before and after stabilization
(samples that were stored for 15 days at 5 °C). Solid fat
contents were calculated as described above and the
diagrams of isosolids were produced using the same
Origin (Microcal, 1999) programme and the Statistica
programme (StatSoft, 2000).
3. Results and discussion
The physiological development of the mangoes used
for MAF extraction was characterized using chemical
analysis of the pulps. The average compositions
(mean ± standard deviation of triplicate analyses) for
unripe mango fruits pulps were: moisture content,
74.57 ± 0.26%; pH values of 3.40 ± 0.23; ash con-
tent, 1.11 ± 0.22; total sugar content, 24.50 ± 1.05%;
titrable acidity, 7.55 ± 0.25; tannins, 0.089 ± 0.008%; and
74 J.A. Sol
ıs-Fuentes, M.C. Dur
an-de-Baz
ua / Bioresource Technology 92 (2004) 71–78

nitrogen-free extract, NFE, 66.73 ± 1.04%, all dry basis, Table 2

except moisture contents. Mature fruits had: moisture Profile of fatty acidsa in the MAF and in a commercial sample of CB
content, 85.07 ± 0.06%; pH values of 4.51 ± 0.09; ash Fatty acid MAF CB
contents, 2.59 ± 0.08%; total sugars, 90.25 ± 0.62%; ti- Palmitic 9.29 24.27
trable acidity, 4.1 ± 0.29; tannins, 0.018 ± 0.001%; and Stearic 39.07 35.10
Oleic 40.81 36.47
FNE, 3.02 ± 0.25%. These values and differences clearly
Linoleic 6.06 2.85
show the inherent characteristics of unripe and mature Linolenic 0.64 0.30
fruits, with ripeness processes increasing sugar contents Arachidic 2.48 1.01
and reducing acidity and tannins. Behenic 0.64 –
Physiological development had no significant effect Lignoceric 0.49 –
No identified 0.52 –
on the weight of each portion of the fruit (peel, pulp,
a
seed and almond). Values found for both unripe and Average of duplicate analyses.
mature fruits were: average weight 298 g, with peel,
16.11%; pulp, 66.51%; seed, 11.69%, and almond, 1980; Talbot, 1995). According to Table 2, CB has a
5.58%. The almond chemical composition is presented little over 60% of saturated fatty acids whereas MAF
in Table 1, for both unripe and mature fruits. Mature has 52% and the ratio among saturated fatty acids to
fruits showed the logical changes in proteins, fats, crude unsaturated ones is, for CB, 1.52 and for MAF, 1.09.
fibres, ashes and tannins, resulting from the cellular These figures are within the reported values for MAF
differentiation and specificity of tissues related with from other mango varieties, and for CB (MdAl
ı et al.,
maturity processes. 1985; Schlichter-Aronhime and Garti, 1988; Sol
ıs-
The fat extracted and purified had a solid consistency Fuentes, 1998).
at environmental temperature of 20–23 °C and pre- Fig. 1 shows the thermograms of fusion for stabilized
sented a cream colour with a characteristic aroma. The MAF and the solidification thermograms of the same fat
physicochemical characteristics of purified samples after being heated until complete fusion and maintained
MAF of mature fruits were: refractive index 1.466, at 90 °C for 10 min. As may be observed, profiles for
acidity value 0.60% (as oleic acid), saponification index phase change S–L and viceversa are relatively simple
189.0 and iodine index of 47.7 (average of two deter- and closely correspond to the similarly relatively simple
minations). fatty acid composition. It is shown that, during fusion,
Table 2 presents the analytical results for the GC-ME there is a low melting point fraction with fusion starting
for MAF and CB (the commercial CB sample had an at a temperature of )17.16 °C and a maximum of 5.32
iodine index of 43.8). In both types of fats, mango al- °C, and a higher melting point fraction, with a less
mond and cocoa, the fatty acids profiles show oleic, opened peak starting at 26.18 °C and presenting a
stearic and palmitic acids. For CB, these three fatty maximum at 40.01 °C, completing the phase change at
acids constitute more than 95% of the total content, around 53.8 °C. The enthalpy change for this fusion
giving this fat a relatively simple acylglyceride compo- process was 115.72 J/g. The cooling process for the
sition. Depending on the origin of the cacao bean, these melting fat starts the solidification step at 14.64 °C and
acylglycerides vary from 80% to 95% (mainly POP and
POS, with some as SOS, where P ¼ palmitic acid,
O ¼ oleic acid, and S ¼ stearic acid). These compositions 2
give unique physicochemical characteristics to this fat,
determining its thermal and phase conduct (Timms, 3
Cooling
4 1

Table 1
Heat Flow

1 3 5
Chemical composition of Mexican mango var. Manila almonds
> >Exo

2
Characteristics (%) Unripe fruits Mature fruits Heating
a
Average ±r Average
r
Moisture contents 32.98 ± 1.63 19.80 ± 1.17
Ashes contents 1.45 ± 0.012 2.08 ± 0.036
Fat contents 5.28 ± 1.13 9.36 ± 1.87
Crude fibre 1.24 ± 0.06 7.58 ± 0.32 4
Proteins 1.80 ± 0.02 4.62 ± 0.13 -80 -60 -40 -20 0 20 40 60 80 100
Tannins 0.04 ± 0.004 1.60 ± 0.034
Cyanides 0.01 ± 0.0007 0.006 ± 0.001
Temperature, °C
FNEb 90.14 ± 0.86 69.25 ± 6.18
Fig. 1. Differential scanning calorimetry crystallization and fusion
a
Mean ± standard deviation of triplicate analyses. curves for stabilized MAF. Refer to Table 3 for (1, 2, 3, 4 and 5)
b
Free nitrogen extract. transition temperatures.
 J.A. Sol
ıs-Fuentes, M.C. Dur
an-de-Baz
ua / Bioresource Technology 92 (2004) 71–78 75

Cooling
3 1
MAF
4 1 6
23 CB

Heat Flow
Heat Flow

1 5

<<Endo
>>Exo

4
1 3 5
2 2 5
Heating

4 4

-60 -40 -20 0 20 40 60 80 100 -80 -60 -40 -20 0 20 40 60 80 100

Temperature, °C Temperature, ºC

Fig. 2. Differential scanning calorimetry crystallization and fusion Fig. 3. Differential scanning calorimetry fusion curves for MAF and
curves for stabilized CB. Refer to Table 3 for (1, 2, 3, 4 and 5) tran- CB without stabilization. Refer Table 3 for (1, 2, 3, 4, 5, and 6)
sition temperatures. transition temperatures.

ends at )24.27 °C, with a crystallization maximum at
10.41 °C. A slight ‘‘shoulder’’ is observed at )9.90 °C,
that might be assumed to be the solidification of the low
fusion point fraction, constituted by triacylglycerides
with unsaturated fatty acids. The value for DHc for this
solidification process was 56.06 J/g.
Fig. 2 shows the profile of fusion for the CB, stabi-
lized under the same conditions as MAF, and the CB
crystallization profile after being completely melted and
maintained at 90 °C for 10 min. In this case, the ob-
served fusion and crystallization profiles are also rela-
tively simple and show a great resemblance with those
for MAF.
The CB also presented the two maxima for fusion.
The first one corresponds to the low fusion point frac-
tion, with a maximum temperature of 11.64 °C and the
second one corresponds to the higher fusion tempera-
ture triacylglycerides with a maximum point at 36.51 °C.
The DHf for the CB analyzed was 128.17 J/g. With re-
spect to crystallization, it starts at 12.71 °C with a
maximum at 6.38 °C, and a DHc of 56.98 J/g.
Similarities in the thermal behaviour of MAF and CB
are also evident in Fig. 3 that shows the fusion profiles
for both fats without stabilization and subjected to im-
mediate heating after solidification with rapid cooling.
Table 3 presents the more representative transition
temperatures in the thermograms, allowing comparison
of the CB with the MAF data.
According to Hagemann (1988), vegetable origin
lipids (fatty acids, acylglycerides, and fats and oils)
present polymorphism, and in general, and with more
frequency, solidify in three different crystalline forms: a,
b0 , and b, with correspondingly higher fusion tempera-
tures. Polymorph a (lowest fusion point) is generally
present after rapid cooling processes from melted fat.
Form b0 , with a higher melting or fusion point than the
previous one, is generated through solidification of fat
under certain conditions of temperature or due to
transition from form a. Polymorph b, the most stable
crystalline form, is produced from the other two forms
by incubating at slightly higher fusion temperatures
than for the a form. Polymorphism for CB has been
widely studied (Lutton, 1957; Schlichter-Aronhime and
Garti, 1988; Loisel et al., 1998) and there is still con-
troversy with respect to the number and types of
forms in which it crystallizes. One of the most accepted

Table 3
Transition point temperatures and crystallization and fusion enthalpies for MAF and CB
Sample DH (J/g) Transition temperature (°C)
1 2 3 4 5 6
Crystallization
MAF 56.06 14.64 10.41 )9.90 )24.27
CB 56.98 12.71 6.38 )15.22 )26.17
Fusion
Stabilized MAF 115.72 )17.16 5.32 26.18 40.01 53.84 n.d.
Non-stabilized MAF 70.12 )15.94 )2.31 2.01 10.62 18.54 42.23
Stabilized CB 128.17 )11.02 11.64 22.80 36.51 52.47 n.d.
Non-stabilized CB 80.02 )3.50 8.08 19.70 23.11 40.18 n.d.
n.d.: Not determined.
76 J.A. Sol
ıs-Fuentes, M.C. Dur
an-de-Baz
ua / Bioresource Technology 92 (2004) 71–78

30
100 Non-stabilized CB
26
SFC = 5
80
Solid Fat Content

10

Temperature °C
Stabilized CB 22
20
60 18 30

Stabilized MAF 14 50
40 60
10 70
Non-stabilized MAF
20
6 80
0 2
0.0 0.2 0.4 0.6 0.8 1.0
-20 -10 0 10 20 30 40 50 60 70 MAF (Weight Fraction)
Temperature, °C
Fig. 5. Isoline diagram of solid fat content (SFC) for mixtures of
Fig. 4. Solid fat content for stabilized and non-stabilized MAF and MAF and CB in non-stabilized samples.
CB.

the polymorphic phases, phase separation, etc.) becomes

proposals is that it may be crystallized in six different
polymorphs (Talbot, 1995). According to this, the fu-
sion thermograms for CB in Figs. 2 and 3 correspond to
forms a and b2 respectively, considering the typification
of Larsson or to forms II and V as a function of the
Wille and Lutton studies (Talbot, 1995).
The thermal behaviour for MAF shown in Figs. 1
and 3 suggests that, following the CB case, it is also a b-
stable fat and different from the so-called b0 -stable fats
such as those obtained from palm, soy or colza oil hy-
drogenation or those known as lauric fats from coconut
or palm seeds.
Fig. 4 shows the MAF and CB samples solids pro-
files, stabilized and without stabilization. Similarities
can be observed in the behaviour of the solid and liquid
phases of the fats due to the temperature effect, when
subjected to stabilization. Once stable, MAF is softer
than CB at lower temperatures and has a harder con-
sistency at higher temperatures, preserving more solids
than CB. This behaviour is probably due to the com-
position differences, since MAF has a somewhat higher
oleic acid content that may have a ‘‘dilution effect’’ for
the saturated fatty acids acylglycerides, slightly reducing
the fusion point. MAF, on the other hand, contains
saturated fatty acids of longer chain, such as behenic
and lignoceric acids, that might be influencing the re-
quirements for higher temperatures for MAF complete
fusion.
a relevant issue for the overall product quality.
Figs. 5 and 6 show the isosolids diagrams for mix-
tures of MAF and CB. This type of diagram links points
of equal contents of solids in the fat, within a compo-
sition-temperature graph, as a way of analyzing the in-
teraction of the components and the fusion profiles for
the mixtures. If the lines are parallel and approximately
horizontal, then the components exhibit good compati-
bility. If this is not the case, with the extreme case oc-
curring when eutectic mixtures are formed (minimum
points), softening effects occur, considering the pure
components, or the formation of higher fusion points
compounds (Birker and Padley, 1987; Talbot, 1995).
In samples without stabilization, the isosolid diagram
of Fig. 5 shows that MAF is slightly softer than CB and
has a more evident softening effect in mixtures when its
contribution to solids is 60–80% by weight. If its con-
tribution to the mixtures is small, the effect is negligible
and if higher, when solids content is small, the effect is
compensated, and MAF requires higher temperatures
than CB to melt. In fact, the profile of fusion for MAF is
wider, melting starting at lower temperatures and phase
changing completely at higher temperatures than CB.
50
44 SFC = 5
10
Temperature °C

As it is well known, fats affect the functional prop- 20

38 30
erties of foods, cosmetics and pharmaceutical products. 50
This can be positive or negative depending upon the 60
32
final product, and may be a multiphase one. The fat 70
phase of a product may be constituted by a fat of single 26
origin or by several fats from different sources and of 80
different compositions, and its stability greatly influ- 20
0.0 0.2 0.4 0.6 0.8 1.0
ences that of the product. The effect of deliberate mixing
MAF (Weight Fraction)
or the migration of the fat in the product on the thermal
behaviour and consistency of the fat phases (resulting Fig. 6. Isoline diagram of solid fat content (SFC) for mixtures of
from the formation of eutectic products, modification of MAF and CB in stabilized samples.
 J.A. Sol
ıs-Fuentes, M.C. Dur
an-de-Baz
ua / Bioresource Technology 92 (2004) 71–78
The diagram shows a lower compatibility between MAF
and CB than between CB and other fats such as Cober-
ine, with isolines more parallel and horizontal during its
mixing with CB (Talbot, 1995). It is, however, much more
compatible than milk fat, lauric fats and hydrogenated
cottonseed oil (Birker and Padley, 1987; Talbot, 1995).
Improved compatibility for the stabilized forms type
b, shown in Fig. 6 is found with no softening effect at
compositions of lower than 20% of MAF, and a harder
mixture for compositions higher than 80% of MAF.
Again, a wide fusion profile for MAF is shown, re-
quiring lower temperatures than CB to have 80% solids
and higher temperatures to melt the last 5% solids of the
pure state fat.
The diagram of the stabilized mixtures shows the
much better compatibility of MAF with CB when
compared with lauric fats and with those coming from
the hydrogenation of vegetable oils.
Based on the results found in this study, mango al-
mond fat from mature fruits of the Manila variety
(MAF), widely produced in Mexico, has physicochem-
ical characteristics and a fatty acids profile close to those
of CB. The MAF curves for solid–liquid phase change
are relatively simple and shows a great resemblance with
those of CB. However, the MAF fusion curve is wider
than that for CB. The SFC MAF profiles show, just as
CB’s, a very steep slope around 30–40 °C. MAF thermal
behaviour with and without stabilization suggests the
presence of a and b crystalline forms, making it a b-
stable fat, just as CB. The isosolid diagrams for mixtures
of MAF and CB showed that there is compatibility
between both fats, even better than that of mixtures of
CB with milk fat, lauric fats, or hydrogenated cotton-
seed oil. Therefore, MAF may be used as a partial
substitute for CB in some of its multiple applications. Its
properties and thermal conduct also make it suitable for
use in mixtures with vegetable oils for specialty fatty
products with specific consistencies and characteristics.
Finally, MAF and its use would give added value to
those residues present in large quantities in the mango
nectars and canned and dried products industries in
tropical countries.
References
AOAC, 1995. In: Cunniff, P. (Ed.), Official Methods of Analysis.
AOAC International, Arlington, Virginia.
Bannon, C.D., Craske, J.D., Hai, N.T., Harper, N.L., O’Rourke,
K.L., 1982. Analysis of fatty acid methyl esters with accuracy and
reliability. J. Chrom. 247, 63–69.
Ben-Gera, I., Kramer, A., 1969. The utilization of food industries
wastes. Adv. Food Res. 17, 77–152.
Bertoli, C., Hosszu-Sackett, K., Melachouris, N., Truifler, H., 1995.
Preparation of butterfat and vegetable butter substitutes. USA
Patent 5395629.
Birker, P.J.M.W.L., Padley, F.B., 1987. Physical properties of fats and
oils. In: Hamilton, R.J., Bhati, A. (Eds.), Recent Advances in
77
Chemistry Technology of Fats and Oils. Elsevier, New York
(Chapter 1).
Dhinigra, S., Kapoor, A., 1985. Nutritive value of mango seed kernel.
J. Sci. Food Agric. 36, 752–756.
El-Zanati, E.M., Zaher, F.A., 1990. Utilization of rice bran oil to
produce an antifoamer. JAOCS 67 (1), 61–63.
Gutcho, M., 1979. Edible oils and fats. Recent Developments. Noyes
Data Co., New Jersey.
Hagemann, J.W., 1988. Thermal behavior and polymorphism of
acylglycerides. In: Garti, N., Sato, K. (Eds.), Crystallization and
Polymorphism of Fats and Fatty Acids, vol. 31. Marcel Dekker,
New York (Chapter 2).
Hendrikse, P.W., Harwood, J.L., Kates, M., 1994. Analytical methods.
In: Gunstone, F.D., Hornwood, J.L. (Eds.), The Lipids Handbook,
second ed. Chapman and Hall Chemical Database, New York.
Horwitzs, W., 1980. Official Methods of Analysis of the Association of
Official Analytical Chemists. AOAC, Washington, DC.
Jim
enez-Berm
udez, M., Silva-Hern
andez, E.R., Sol
ıs-Fuentes, J.A.,
Dur
an-de-Baz
ua, M.C., 1995. La grasa de semilla de mango como
posible sustituto de la manteca de cacao. In: I Congreso
Iberoamericano de Ingenier
ıa de los Alimentos, November 4–7,
1995. Campinas, S.P. Brazil.
Khosla, P., Sundram, K., 1996. Effects of dietary fatty acid compo-
sition on plasma cholesterol. Prog. Lipid Res. 35 (2), 93–132.
Kloek, W., Walstra, P., Van Vliet, T., 2000. Crystallization kinetics of
fully hydrogenated palm oil in sunflower oil mixtures. JAOCS 77
(4), 389–398.
Lakshminarayana, G., Chandrasekhara-Rao, T., Ramalingaswamy,
P.A., 1983. Varietal variations in content characteristics and
composition of mango seed and fat. JAOCS 60 (1), 88–89.
Lambelet, P., Raemy, A., 1983. Isosolid diagrams of fat from thermal
analysis data. JAOCS 60 (4), 845–847.
Loisel, C., Keller, G., Lecq, G., Bourgaux, C., Ollivon, M., 1998.
Phase transitions and polymorphism of cocoa butter. JAOCS 75
(4), 425–439.
Lutton, E.S., 1957. On the configuration of cocoa butter. JAOCS 34,
521.
Maurya, K.R., Chawdhary, D.N., 1977. Utilization of mango wastes.
Ind. Food Packer 31, 83–84.
Mehlenbacher, V.C., 1970. An
alisis de grasas y aceites. Ediciones
Urmo, Bilbao.
MdAl
ı, A., Gafur, M.A., Rahman, M.S., Ahmed, G.M., 1985.
Variations in fat content and lipid class composition in ten
different mango varieties. JAOCS 62 (3), 520–523.
MdAli, A.R., Dimick, P.S., 1994. Thermal analysis of palm mid-
fraction, cocoa butter and milk fat blends by differential scanning
calorimetry. JAOCS 71 (3), 299–302.
Microcal, 1999. Origin 6.0. Microcal Software Inc., Northampton, MA.
Narashima-Char, B.L., Reddy, B.R., Thirumala-Rao, S.D., 1977.
Processing mango stones for fat. JAOCS 54, 494–495.
O’Brien, R.D., 1998. Fat and Oils. Technomic Pub. Co., Lancaster,
Penn.
Schlichter-Aronhime, J., Garti, N., 1988. Solidification and polymor-
phism in cocoa butter and the blooming problems. In: Garti, N.,
Sato, K. (Eds.), Crystallization and Polymorphism of Fats and
Fatty Acids, vol. 31. Marcel Dekker, New York (Chapter 9).
Sol
ıs-Fuentes, J.A., 1998. El aprovechamiento de residuos s
olidos y
l
ıquidos de la agroindustria alimentaria, cinco ejemplos. M.S.
Thesis. Facultad de Qu
ımica, UNAM. M
exico, D.F.
Sol
ıs-Fuentes, J.A., Dur
an-de-Baz
ua, M.C., 2001. Los
acidos grasos
trans en los alimentos y algunos efectos de su consumo en la salud.
Revista Ciencia (M
exico) 52 (1–2), 4–14.
Southgate, D.A.T., 1991. Determination of Food Carbohydrates.
Elsevier Applied Science, London.
StatSoft, 2000. Statistica for Windows. StatSoft Inc., Tulsa.
Timms, R.E., 1980. The phase behavior of mixtures of cocoa butter
and milk fat. Lebensm. Wiss. Technol. 13, 61–65.
78 J.A. Sol
ıs-Fuentes, M.C. Dur
an-de-Baz
ua / Bioresource Technology 92 (2004) 71–78

Talbot, G., 1995. Fat eutectics and crystallisation. In: Beckett, S.T.
(Ed.), Physico-Chemical Aspects of Food Processing. Blackie Aca-
demic and Professional. Chapman and Hall, London (Chapter 7).
Toro-V
azquez, J.F., Brice~
no-Montelongo, M., Dibildox-Alvarado, E.,
Char
o-Alonso, M., Reyes-Hern
andez, J., 2000. Crystallization
kinetics of palm stearin in blends with sesame seed oil. JAOCS 77
(3), 297–310.
Williams, S.D., Ransom-Painter, K.L., Hartel, R.W., 1997. Mixtures
of palm kernel oil with cocoa butter and milk fat in compound
coatings. JAOCS 74 (45), 357–366.