Investigación preliminar de la influencia del
trata-miento térmico y tiempo de almacenamiento
sobre el aroma de habas (Vicia faba)
Preliminary investigation on the influence of heat treatment and
storage time on the aroma of faba beans (Vicia faba)
Recibido: enero 15 de 2016 | Revisado: febrero 17 de 2016 |
Aceptado: abril 26 de 2016
Jose Ramos-Díaz1,2, Sonja Suvanto1, Elina Forsten1, Laila Seppa1
Ab s t r act
Faba bean (Vicia faba) is a legume available in
Finland that is used in human and animal
consumption due to its high content of protein
and fiber. Unlike other legumes, little is known
about the aroma volatiles developed by faba
beans in storage, and the descriptors associated to
it. Suspensions containing heat treated (H) and
non-heat treated (NH) faba bean samples were
stored at 4 °C for 0, 7 and 14 days, and evaluated
with descriptive analysis. The training sessions
encouraged the elicitation of aroma descriptors
linked to faba bean. The final attributes were:
fresh pea, musty, bitter, grassy, yeast-like and
total intensity. Faba bean suspensions (H and
NH) were also analysed for volatile compounds
using SPME-GC-MS. The results indicated that
the intensity of total intensity, yeast-like, bitter
and musty increased at the longest storage time
regardless of the heat treatment. In contrast, fresh
pea and grassy appeared to reach a peak at 7-day
storage. Statistical results suggested heat
treatment had a much lesser effect on aroma
development than storage time. The analysis of
volatile compounds showed that 2-butanone-3-
hydroxy and 3-methyl-1-butanol had the most
prominent peak areas at 7 and 14-day storage
(particularly for NH). H and NH samples were
associated to almost the same volatile compounds
until the 7 day and linked to distinct compounds
by the end of the storage. It is believed that fresh
pea and grassy are strongly associated to ketones
present in H and NH samples at 7-day storage
while musty, bitter and yeast-like are more
associated to alcohols and ketones present in NH
samples at 14-day storage.
Keywords: Faba beans, aroma, volatile
compounds, storage, gas chromatography
Re su m e n
Haba (Vicia faba) es una leguminosa disponible en Finlandia que
se utiliza en la alimentación humana y animal debido a su alto
contenido de proteína y fibra. A diferencia de otras legumbres,
poco se sabe acerca de las sustancias volátiles que se producen
durante el almacenamiento y los descriptores de aroma asociados a
las mismas. durante el almacenamiento y los descriptores de aroma
asociados a las mismas. Las suspensiones fueron elaboradas en
base a harina de habas. Muestras fueron sometidas a tratamiento
térmico (T) y mantenidas sin tratamiento térmico (ST). Estas
fueron almacenadas a 4 ° C durante 0, 7 y 14 días. Los panelistas
que participaron en el estudio sensorial fueron ocho mujeres (n =
8) entre 25 y 55 años de edad. Las sesiones de entrenamiento
alentaron a la obtención de los descriptores aromáticos vinculados
al haba: ‘guisantes frescos’, ‘rancio’, ‘amargo’, ‘herboso’,
‘levadura’ y la intensidad total. El análisis sensorial se llevó a cabo
por duplicado con una recesión de 30 minutos en el medio.
También se analizaron las suspensiones de habas (T y ST) para
compuestos volátiles utilizando SPME-GC-MS. Los resultados
indicaron que la intensidad total, levadura, amargo y rancio
aumentó durante almacenamiento prolongado, independientemente
del tratamiento térmico. En contraste, el aroma a guisante fresco y
herboso parecían alcanzar un pico en el almacenamiento 7mo día.
Los resultados estadísticos sugirieron que el tratamiento térmico
tuvo un efecto mucho menor en el desarrollo del aroma que el
tiempo de almacenamiento. El análisis de compuestos volátiles
mostró que 2-butanona-3-hidroxi y 3-metil-1-butanol tenían las
áreas de los picos más prominentes al 7mo y 14vo día de
almacenamiento (especialmente para ST). Muestras T y ST se
asociaron a casi los mismos compuestos volátiles hasta el 7mo día,
y después se vincularon a distintos compuestos al final del
almacenamiento. Se cree que los aromas a guisante fresco y
herboso están fuertemente asociados a cetonas presentes en
muestras T, y muestras ST al 7mo día de almacenamiento,
mientras que rancio, amargo y levadura están más asociados a los
alcoholes y cetonas presentes en las muestras ST al 14vo día de
almacenamiento.
Palabras clave: habas, aroma, compuestos volátiles,
almacenamiento, cromatografía de gases
1 Department of Food and Environmental Sciences, PO Box 66 (Agnes Sjöbergin katu 2) FI-00014 University of Helsinki, Finland
2 E-mail: jose.ramosdiaz@helsinki.fi
| Cátedra Villarreal | Lima, perú | V. 4 | N. 1 | PP. 11-24 | enero-junio | 2016 | issn 2310-4767 11
Jose Ramos-Díaz, Sonja Suvanto, Elina Forsten, Laila Seppa
INTRODUCTION
Legumes are nutritious food alter-
natives for humans (Tiwari, Gowen, &
McKenna, 2011). They contain essential
amino acids, fatty acids, vitamins, min-
erals, carbohydrates and fiber. Legumes
include a large number of different
beans, which are able to grow in diffi-
cult conditions. Its advantage over oth-er
plants is their ability to fix nitrogen from
the soil (formation of ammonia from
molecular nitrogen). Faba beans (Vicia
faba) are one of the crops that are better
adapted to the Nordic climate (Peltonen-
Sainio and Niemi, 2012).
Generally, faba beans are used in
whole-grain products, but their tech-
nological use increased considerably
since early 2000s (Makri, Papalamprou,
& Doxastakis, 2005). The use of vegeta-
ble proteins as functional components in
food is limited (almost entirely) to
soybeans, whose nutritional content is
similar to any other legume. Promoting
the efficient use of alternative legumes,
and studying their technological and
sensory properties have become a sci-
entific challenge.
The particular aroma of legumes
limits their use in the Western world
(Vara-Ubol, E. Chambers, & D. Cham-
bers, 2004). For instance, the flavor gen-
erated by the oxidation of unsaturated
fatty acids during processing and stor-
age, and/or the degradation of amino
acids during heat treatment result in the
release of volatile compounds of poten-
tial unpleasant aroma (Rodríguez-Ber-
naldo De Quirós, López-Hernández,
González-Castro, De La Cruz-García, &
Simal-Lozano, 2000). Bott and Cham-
bers (2006) found two separate volatile
compounds that have no beany flavor,
but as a whole generated the character-
istic faba bean aroma. Vara-Ubol et al.
(2004) had a trained panel (n = 5) to
evaluate the aroma and taste of differ-ent
varieties of canned and dried faba beans.
Aroma and taste as a whole were
described with the words: ‘stale’, ‘hu-
midity’, ‘bitter’, ‘green’, ‘nut’ and
‘starch’. The structure was described as
‘flour-like’. Certain volatile compounds
such as alcohols, ketones, aldehydes, and
pyrazine contributed to aroma and taste of
the samples. The authors also found that
the concentration of volatiles was
relevant. Many compounds, produced in
small amounts, contributed to the aroma
or taste as a whole, but such ef-fect was
lost at higher concentrations.
As descriptors for faba beans are
hardly found in the literature, the aim
of this work was to create a descriptive
profile of the aroma for faba bean sus-
pensions (in water), and test the effect
of heat treatment and storage time on
the aroma and intensity. The correla-
tion between volatile aromatic com-
pounds present in the samples, and the
aromas described by the panellists will
also be studied. The work was done in
collaboration with the Department of
Food and Environmental Sciences at
the University of Helsinki.
METHOD
Materials
Faba beans (Vicia Faba var. Kontu
2012) used in this study were harvest-ed
from the teaching and research farm of
Helsinki University in Viikki. Faba
beans were divided into two groups:
heat treated (H) and non-heated treat-ed
(NH) samples. The H samples went
through moisture adjustment to 10%
followed by 30-min stabilization time.
These samples were heated at 140 °C for
20 min and then stored overnight at 40
°C. Eventually, H and NH samples were
dehulled, milled and stored at room
temperature prior to the preparation of
12 | Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 |
Investigación preliminar de la influencia del tratamiento térmico y tiempo de almacenamiento sobre el aroma de habas (Vicia faba)
suspensions. These suspensions were
elaborated by placing 20 g of H or NH
samples in a 150-ml vial followed by the
addition of 100 ml water. This mixture
was first stabilized for 30 min and then
centrifuged for 15 min at 10000 rpm.
The supernatant was poured in a 200-ml
flask, sealed and stored in darkness at 4
°C for 0, 7 and 14 days. Suspen-sions
were prepared in the Cereal Tech-nology
laboratory at University of Hel-sinki.
The process is shown in Figure 1.
Figure 1. Flow chart for the preparation of faba bean suspensions.
Methods
Training of panellists
The panellists involved in the study were
8 women (n = 8), aged between 25 and
55 years old (staff and students). Most of
them had taken part in sensory testing
before, and were familiar with faba bean.
The training sessions were held in
Finnish at the food sensory laboratory of
Department of Food and Environmental
Sciences. This study followed the
approved ethical require-ments of the
University of Helsinki, and an informed
consent was signed by the participating
panellists. Two training
sessions (2.5 h in total) were arranged
prior to the sensory test. During the
first session (90 min), panellists were
presented with a series of hulled faba
bean suspensions as references (Ta-
ble 1). They described the samples in-
dependently and then discussed with
the rest of the panel. A list of descrip-
tors was presented (Appendix 1) so as
to help panellists to form a concept
around the aroma characteristics of H
and NH faba bean suspensions. At
first, panellists came up with a large
num-ber of descriptors (Appendix 2)
which eventually narrowed down to 6
(Table 2), including ‘total intensity’.
Table 1
List of reference samples used for both training days
Reference sample Manufacturer Sample size
Container
7-day H hulled faba bean susp.
Food Tech. Lab. at UH
15 ml
14-day H hulled faba bean susp.
Food Tech. Lab. at UH
15 ml
7-day NH hulled faba bean susp.
Food Tech. Lab. at UH
15 ml
14-day NH hulled faba bean susp.
Food Tech. Lab.at UH
15 ml
Laboratory
bottles
| Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 | 13
Jose Ramos-Díaz, Sonja Suvanto, Elina Forsten, Laila Seppa
Yeast
Suomen Hiiva
50 g
Soil with water
Food Tech. Lab.at UH
50 g
Fresh green peas
Apetit
100 g
Soaked peas
Food Tech. Lab.at UH
150 g
Almonds
Kluth
50 g
Cashew nuts
Kluth
50 g
Brazilian nuts
Kluth
50 g
Hazelnuts
Kluth
50 g
Hexanal
Food Tech. Lab.at UH
50 g
Fresh mushroom
X-tra
50 g
Dried mushroom
Self-collected
50 g
Dried black chanterelle
Self-collected
50 g
Dried funnel chanterelle
Self-collected
50 g
Porcelain con-
tainers
Note: Sample containers were covered by a plastic foil.
During the second training (60 min), the
actual faba bean suspensions were
revealed as such so that panellists could
attempt to describe their odours with the
help of the descriptors chosen by con-
sensus during the first training. There
were, however, changes in the vocabu-
lary such as the addition of ‘yeast-like’
and the withdrawal of ‘toasty’ and ‘pea-
nut’. The final set of the descriptors as-
sociated to the odour of H and NH faba
bean suspensions are detailed in Table 2.
Table 2
Descriptors chosen by consensus during the first and second training
First training – Ensimmäinen koulutus
Fresh pea – tuorehernemäinen
Musty – ummehtunut
Bitter – kitkerä
Toasty – paahteinen
Grassy – ruohomainen
Peanut – pähkinäinen
Total intensity – Kokonaisvoimakkuus
Note: The training was originally conducted in Finnish language.
14 | Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 |
Investigación preliminar de la influencia del tratamiento térmico y tiempo de almacenamiento sobre el aroma de habas (Vicia faba)
Sensory profiling
The data collection method was gener-
ic descriptive analysis as described by
Lawless and Heymann (2010). The de-
scriptors were introduced in Fizz® soft-
ware (version 2, Biosystèmes, Couter-
non, France) for computer-based
testing. A 15 cm magnitude line scale
with indented end-anchors lines on ei-
ther side was used. For ‘total intensity’,
labels endpoints were ‘not strong at all’
(ei lainkaan voimakas) and ‘extreme-ly
strong’ (erittäin voimakas) while for
the rest of the attributes, labels end-
points were ‘not at all’ (ei lainkaan)
and ‘extremely’ (erittäin). Coded
samples were randomized across trials.
Panel-lists were first presented with
one set of samples (3 H samples and 3
NH sam-ples, corresponding each to 0,
7 and 14-day storage) and, after a 30-
min re-cession, with a second set of
samples (duplicate).
Analysis of volatile compounds
Faba bean suspensions (H and NH) at
different storage times (0, 7 and 14
days) were analysed for volatile com-
pounds by using SPME-GC-MS. Sol-
id-phase micro extraction (SPME) was
tested with Divinylbenzene / carboxen
/ polydimethylsiloxane fiber (DVB/
CAR/PDMS 50/30 μm) (Supelso, USA)
by using an auto sampler system (com-
biPAL, CTC Analytics). The sample (2
ml) was incubated in a 20 ml vial with
pierceable silicon/Teflon disks in the
cap. The fibers were directly desorbed at
250 °C (DVB/CAR/PDMS) for 10 min
in the injection port of GC. The GC-MS
system consisted of a GC (HP 6890 se-
ries) linked to a mass selective detector
(MS) (Agilent 5973 Network). The GC
was equipped with a SPB-624 column
(Supelco, Bellefont PA, USA) (30 m x
0.25 mm i.d., 1.4 μm film thickness).
The following parameters were select-
ed for the SPME-GC-MS method:
fiber DVB/CAR/PDMS 50/30 μm;
incuba-tion at 50 °C for 20 min;
extraction at 50 °C for 30 min. The
GC programme started at 35 °C for 5
min, increased 3 °C/min until 150 °C,
held for 1 min, in-creased 6 °C/min
until 200 °C and held for 5 min.
Statistical analysis
The resulting data was statistically an-
alysed by using a three-way repeat-ed-
measures ANOVA in SPSS (SPSS
18.0, PASW Statistics, Chicago, IL,
USA). Factors were heat treatment
(number of factors 2), storage time
(3) and replicates (2). Fisher’s LSD
test with significance level of 5% was
used for pairwise comparisons.
Principal component analysis (PCA)
plots were used to identify the most
important di-rections of variability of
different sam-ples in a multivariate
data matrix (The Unscrambler v9.7;
CAMO Software AS, Oslo, Norway);
correlations among de-scriptors and
volatile compounds were identified.
RESULTS
Aroma characteristics of the
suspensions and intensity
Aroma intensity increases with in-
creasing length of storage, both in H and
NH samples. Similarly, yeast-like, musty
and bitter aromas were perceived more
intensely at longer storage peri-ods.
Grassy and fresh pea reached their
greatest perceivable intensity at 7-day
storage. Means and standard deviations
are shown in Figure 2, and the effect of
heat treatment and storage time on aro-
mas in Figure 3. The standard deviation
was relatively high in all results. It was
observed that H and NH samples had
similar aroma to fresh peas, yeast-like
and total intensity before storage (0-day
| Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 | 15
Jose Ramos-Díaz, Sonja Suvanto, Elina Forsten, Laila Seppa
storage). The NH samples stored for 0
and 7 days possessed a stronger faba
bean aroma than H samples. Regard-
less of the treatment, samples stored
for 7 days elicited the strongest grassy
aroma. After 14-day storage, H and NH
samples possessed yeast-like, musty and
bitter aroma. The same observa-tions
can be made based on PCA.
Figure 2. The averages of the every perceived aroma characteristics of faba bean
suspensions sub-jected to different heat treatments and storage times. ST = No heat
treatment; T = Heat treatment; 0, 7, 14 days of storage. Error bars show ±1 SD. n=8
16 | Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 |
Investigación preliminar de la influencia del tratamiento térmico y tiempo de almacenamiento sobre el aroma de habas (Vicia faba)
Figure 3. Effect of heat treatment and storage time on perceived aromas. The experiment
was con-ducted in duplicate.
*n = 8.
The heat treatment and storage time
affected significantly [main effect of
heat treatment, F (1, 7) = 13.1, p=0.008;
main effect of storage time, F (2, 6) =
28.4, p=0.001] the perceived total in-
tensity. The heat treatment and storage
time had also a considerable effect on
the aroma of fresh pea [main effect of
heat treatment, F (1, 7) = 6.3, p=0.04;
main effect of storage time, F (2, 6) =
6.4; p=0.03] and grassy [main effect of
heat treatment, F (1, 7) = 17, p=0.004;
main effect of storage time, F (2, 6) =
6.4; p=0.03]. It seems that storage time
increased distinctively the yeast-like
[main effect of storage time, F (2, 6) =
20.9; p=0.002] and bitter aroma [main
effect of storage time, F (2, 6) = 12.3;
p=0.007]. Regardless of the treatment,
the musty aroma increased consider-
ably between 7 and 14 days of storage
(pairwise comparison; p=0.015).
The replicate of the evaluation had no
significant effect on the aroma intensity
or any individual aroma. This means
that, in general, panellists evaluated
samples reliably and accurately. How-
ever, storage time and replicate had a
significant interaction effect [F (2, 6) =
6.4; p=0.03] on musty aroma. Besides,
the interaction of heat treatment and
storage time had a significant effect [F
(2, 6) = 9.1; p=0.01] on yeast-like aro-
ma. This means that the effect of storage
time on the yeast-like aroma changed
depending on the heat treatment.
| Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 | 17
Jose Ramos-Díaz, Sonja Suvanto, Elina Forsten, Laila Seppa
Principal component analysis
Based on the PCA, two principal com-
ponents explained 97% of the observed
variation in the aroma characteristics of
the samples (Figure 4). The first prin-
cipal component explained 66% of the
variation and the second 31%. Samples
with different storage times (0, 7 and 14
days) showed a higher variation along
the first principal component (66%
variation). Conversely, the heat
treated samples showed no significant
differ-ences, and these appear to be
better ex-plained by the second main
component (31% variation).
Figure 4. PCA plot for the perceived intensity of aromas. NH = No heat treatment; H =
Heat treat-ment; 0, 7, 14 days of storage.
Analysis of volatile compounds
The volatile compounds found in the
samples are presented in the PCA bi-
plot (Figure 5). There, the aroma com-
pounds of the samples (surface areas of
the chromatographic peaks) and the
averages of the detected intensity of
the aromas are combined. The vol-atile
compounds of greater quantity are
listed in Table 3. In Figure 5, it was
observed that the aroma of bitterness
and yeast-like correlated with 3-meth-
yl-1-butanol. The concentration of this
compound in H and NH samples, and
stored for up to 14 days, was relatively
low compared to other compounds
(Ta-ble 3). Moreover, H samples
stored for 7 and 14 days reached the
highest levels of 3-methyl-1-butanol.
18 | Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 |
Investigación preliminar de la influencia del tratamiento térmico y tiempo de almacenamiento sobre el aroma de habas (Vicia faba)
Figure 5. PCA plot where volatile compounds (identified through gas chromatography)
and the perceived intensity of aromas are combined. NH = No heat treatment; H = Heat
treatment; 0, 7, 14 days of storage.
Table 3
Major volatile compounds in the suspensions of faba beans
NH/0
NH/7
NH/14
H/0
H/7
H/14
2-methylfu-
1-hexanol
2-butano-
2-methylfu-
3-methyl-1-bu-
3-methyl-1-bu-
ran
(19 946 670)
ne-3-hydroxy
ran
tanol
tanol
(4 115 040)
(41 782 412)
(3 707 030)
(22 769 940)
(25 393 018)
2-butano-
3-methyl-1-bu-
1-hexanol
Acetic acid
ne-3-hydroxy
tanol
(17 747 355)
(22 446 826)
(18 463 582)
(40 219 860)
3-methyl-1-bu-
1-hexanol
2-butano-
2-butano-
tanol
(17 516 052)
ne-3-hydroxy
ne-3-hydroxy
(15 313 897)
(17 417 733)
(12 899 695)
2-methylfuran
2-butanone
3-methylbutanal
1-hexanol
(4 223 931)
(5 571 265)
(10 864 550)
(10 701 280)
2-methylfuran
(4 747 816)
Note: The numbers in parentheses represent the average of the areas under the chromatographic
peaks. NH= No heat treatment; H = Heat treatment; 0, 7, 14 days of storage.
| Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 | 19
Jose Ramos-Díaz, Sonja Suvanto, Elina Forsten, Laila Seppa
Content of 3-methyl-1-butanol
seems to correlate with the overall fla-
vor intensity. Based on Figure 5, the
overall flavor intensity can be related
to the concentration of compounds
such 2-butanone-3-hydroxide, hexanal
and hexanol-1. The concentration of 2-
bu-tanone-3-hydroxide was higher in
the NH samples, stored for 14 days; 2-
bu-tanone-3-hydroxido also had a high
correlation with the total intensity of
aromas. The second largest concentra-
tion of 2-butanone-3-hydroxide cor-
responds to NH samples, stored for 7
days, and the third largest concentra-
tion of this compound corresponds to
H samples, stored for 7 days (slightly
higher than H samples stored for 14
days). The concentration of 1-hexanol
was the greatest in NH samples, stored
for 7 days; these samples are character-
ized by a strong aroma of fresh peas. In
H samples, the concentration of 1-hex-
anol was low compared to NH
samples. Given the intensity of
grassiness and the high concentrations
of 1-hexanol, it is reasonable to think
that this com-pound is somehow
related to perceived grassiness.
Based on Figure 5, the intensity of
grassy aroma correlated with various
compounds such as 2-heptanone, oc-
tane, 2-octanone, 3-methyl butanal, 2-
nonanol, and 2-methyl butanal, but
only the concentration of 3-methyl-
butanal has been observed in H sam-
ples, stored for 7 days. The aroma of
fresh pea was strongly associated with
2-pentylfuran, but this compound was
found in low concentrations in every
sample. Interestingly, the results of sen-
sory evaluation showed that the aro-ma
of fresh peas was one of the most
intense. This suggests that there is no
direct correlation between the aroma of
fresh peas and the 2-pentylfuran, or that
the compound causes a strong
aroma even at low concentrations.
Fresh samples presented aromas of
low intensity and, generally, very low
concentration of volatile compounds.
2-methylfuran had the highest concen-
tration in fresh samples, but was also
detected in samples stored for 7 days.
At 14-day storage, 2-butanone
and 2,3-butane were detected in NH
samples while acetic and butyr-ic
acid were detected in H samples.
DISCUSSION
The panellists were able to differentiate
between the following aromatic descrip-
tors: Fresh peas, yeast-like, musty, bitter
and grassy. More reliable results would
have been obtained with a greater num-
ber of panellists and longer training.
Despite this, panellists’ performance
was acceptable. The effect of the inter-
action between replication and storage
time on musty aroma was significant (p
= .03), which may indicate some dis-
crepancy among panellists (Kälviäinen,
Roininen, & Appelbye, 2005). Panellists
may not have understood attributes same
way, or the samples were not suf-
ficiently distinguishable. It is believed
that the interaction with replicate and the
large standard deviations may have
resulted from insufficient training.
As mentioned in the introduction, the
characteristic aroma of the beans and the
generation of volatile com-pounds have
been a topic of some study (Vara-Ubol
et al., 2004; Bott & Cham-bers, 2006).
The head-space analyses of H and NH
samples, stored for up to 14 days,
allowed the identification of vari-ous
oxidation products such as hexanal, 1-
hexanol, 2-heptanone, 2-pentylfu-ran, 1-
octen- 3-ol, and hexanoic acid. From
these, the concentration of 1-hex-anol
was high in most samples. In addi-tion
to 1-hexanol, the concentration of
20 | Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 |
Investigación preliminar de la influencia del tratamiento térmico y tiempo de almacenamiento sobre el aroma de habas (Vicia faba)
2-butanone-3-hydroxy and 3-3-meth-yl-
1-butanol was considerable. In lit-
erature, at least 3-methyl-1-butanol has
been found to cause, in conjunc-tion
with hexanal, aromas identified as
“peas”, “bad boil ‘,’ horseradish ‘. These
compounds together with 1-octen-3-ol
seemed to cause flavors identified as
‘beans’, ‘peas’, ‘bitter’ and ‘dill’ (Bott
& Chambers, 2006). The correlation be-
tween the perceived intensities of var-
ious aromas and volatile compounds
suggests the positive effect of some
compounds of low concentration on the
formation of aromas. Therefore, the
combined impact of volatiles could de-
termine the occurrence of a particular
aroma detected by the panel.
The total aroma intensity appeared to
be largely affected by the concentration
of compounds such as 3-methyl-1-bu-
tanol, 2-butanone and 3-hydroxy 1-
hexanol. The possible interactions be-
tween volatiles make it hard to suggest a
direct relationship between concen-
tration and intensity. For example, Bott
and Chambers (2006) noted that the
most intense aroma of pea formed from
a combination of hexanal and 1-octen-3-
ol. The same combination may have
occurred in this study, even if the con-
centrations of these compounds were
low. In order to obtain reliable results,
these compounds should be studied in
aqueous suspensions of beans, and then
identified their role in the formation of
specific aromas.
According to the study conducted by
Wang, Dou, Macura, Durance and Na-
kai (1998), alcohols, aldehydes, hexanol
and hexanal are the main compounds
associated with the aroma of green bean
in many soy products. Wang et al.
(1998) also found that hexanol, hexanal
and the lipoxygenase enzyme present a
positive linear correlation. In the pres-
ent study, several volatile compounds
such as 1-hexanol, and the total aroma
intensity were mostly detected in NH
samples. This suggests that the heat
treatment may have an inhibitory effect
on the enzymatic activity and thus, sup-
pressed volatile production. Due to the
presence of amino acids, volatile com-
pounds with amide groups could be re-
leased during heat treatment, leading to
the formation of aromas during storage
(Rodríguez-Bernaldo De Quirós et al.,
2000). The storage time intensified the
formation of aromas in suspension, in-
dicating possible enzyme activity and/
or oxidation of lipids. Also, it is note-
worthy that the suspensions were pre-
pared with distilled water, so all bac-
teria and yeast present in faba beans
could have caused lipid oxidation and
fermentation, particularly in samples
without heat treatment. Thus, it is rea-
sonable to think that the aromas per-
ceived from suspensions stored for 14
days, such as yeast-like or bitter, may
well be off-odours induced by spoilage.
CONCLUSIONS
This preliminary study showed that
aroma of faba bean suspensions was
mostly affected by storage time, while
heat treatment had a lesser effect. An
interaction effect between replicate and
storage time showed that panellists may
not have been thoroughly trained for
aroma detection, or samples were indis-
tinctively similar. Despite this, replicate
had (per se) no distinct effect on the re-
sults. Regardless of the heat treatment,
suspensions stored for 7 days were
strongly associated with the aroma of
fresh pea and grassy while those stored
for 14days were associated with the
aroma of musty, bitter and yeast-like. In
terms of volatile compounds, major
differences were not observed up to 7
days. However, heat treated suspensions
stored for up to 14 days had a consider-
ably higher content of e.g. acetic acid,
| Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 | 21
Jose Ramos-Díaz, Sonja Suvanto, Elina Forsten, Laila Seppa
and much lower content of e.g. 3-meth-
yl-1-butanol compared to the non-heat
treated suspensions. Enzymatic activity
in non-heat treated suspensions might
have led to production of distinct vol-
atile compounds with, apparently, the
same effect on the aroma profile as heat
treated suspensions.
ACKNOWLEDGEMENTS
We wish to thank M.Sc. Marjo Pulkki-
nen (Ph.D. student at University of Hel-
sinki) for her assistance during sam-ple
preparation and analysis of volatile
compounds. M.Sc. Leenamaria Järvin-
en is also thanked for providing relevant
literature in support of our research. Dr.
Kevin Deegan and Technician Jut-ta
Varis from University of Helsinki are
thanked for their technical help during
training and sensory evaluation. This
preliminary study was conducted in
cooperation between the Food Sensory
group, led by Prof. Hely Tuorila, and the
Food Chemistry group, led by Prof. Vi-
eno Piironen.
REFERENCES
Bott, L. and Chambers, E. (2006).
Sen-sory characteristics of
combina-tions of chemicals
potentially asso-ciated with
beany aroma in foods. Journal of
Sensory Studies, 21(3), 308-21
Kälviäinen, N., Roininen, K. & Ap-
pelbye, U. (2005). Raadin valinta,
harjaannuttaminen ja seuranta. H.
Tuorila & U. Appelbye (Eds.),
Elin-tarvikkeiden aistinvaraiset
tutki-musmenetelmät (pp 157-
174). Hel-sinki: Gaudeamus.
Lawless, H.T., Heymann, H. (2010).
Sensory Evaluation of Food: Prin-
ciples and Practices. New York, NY:
Springer-Verlag.
Makri, E., Papalamprou, E., Doxas-
takis, G. (2005). Study of functional
properties of seed storage proteins
from indigenous European legume
crops (lupin, pea, broad bean) in
admixture with polysaccharides.
Food Hydrocolloid, 19, 583-594.
Peltonen-Sainio, P. and Niemi, J.K.
(2012). Protein crop production at
the northern margin of farming: to
boost, or not to boost. Agricultural
and Food Science, 21(4), 370-383
Rodríguez-Bernaldo
De
Quirós,
A.,
López-Hernández,
J.,
González-Castro, M., De La Cruz-
García, C. and Simal-Lozano, J.
(2000).
Comparison of
volatile
components in raw and cooked
green beans by GC-MS using dy-
namic
headspace
sampling and
microwave desorption.
European Food Research and
Technology, 210, 226-30.
Tiwari, B., Gowen, A. and
McKenna, B. (2011). Pulse
Foods: Processing, Quality and
Nutraceutical Appli-cations (1st
ed.). USA: Academic Press.
Vara-Ubol S, Chambers, E., Chambers,
D.H. (2004). Sensory characteris-
tics of chemical compounds poten-
tially associated with beany aroma
in foods. Journal of Sensory
Studies, 19(1), 15-26.
Wang, Z.H., Dou, J., Macura, D., Du-
rance, T.D. and Nakai, S. (1998).
Solid phase extraction for GC
anal-ysis of beany flavours in
soymilk. Food Research
International, 30(7), 503-511.
22 | Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 |
Investigación preliminar de la influencia del tratamiento térmico y tiempo de almacenamiento sobre el aroma de habas (Vicia faba)
APPENDIX
Appendix 1. List of descriptors suggested by the panel leader during the first training.
The training was originally conducted in Finnish language*.
Descriptors [ENG]
Descriptores [SPA]
Kuvaajat [FIN*]
Almond sour
Almendra amarga
Mantelimainen hapan
Beany
(similar a) frijol
Papumainen
Bitter
Amargo
Kitkerä
Brown
Marrón
Ruskea
Dusty
Polvoso
Pölyinen
Fat
Graso
Rasvainen
Fungous
Fungoso
Sienimäinen
Grassy
(similar a) césped
Ruohomainen
Green
Verde
Vihreä
Moldy
Mohoso
Homeinen
Musty
Rancio
Ummehtunut
Nutty
(similar a) nuez
Pähkinäinen
Oily
Aceitoso
Öljymäinen
Pod
Guisantes
Herneenpalko
Soil-like
(similar a) suelo
Maamainen
Stale
Pan rancio
Tunkkainen
Starch
Almidón
Tärkkelys
Appendix 2. Total set of descriptors proposed by the panel during the first training.
The training was originally conducted in Finnish language.
Descriptors [ENG]
Descriptores [SPA]
Kuvaajat [FIN]
Bitter
Amargo
Kitkerä
Cheese-like
(similar a) queso
Juustoinen
Fresh
Fresco
Raikas
Fresh pea
Guisantes frescos
Tuore hernemäinen
Grassy
(similar a) césped
Ruohomainen
Green
Verde
Vihreä
Liquorice-like
(similar a) Regaliz
Lakritsimainen
Mushroom-like
(similar a) Hongos co-
Herkkusienimäinen
mestibles
Musty
Rancio
Ummehtunut
| Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 | 23
Nutty
(similar a) nuez
Pähkinäinen
Oily
Aceitoso
Öljymäinen
Peel-like
(similar a) cáscaras de
Kuoren aromi
fruta
Petrol-like
(similar a) gasolina
Bensa
Sesame
Ajonjolí
Seesami
Soil-like
(similar a) suelo
Maamainen
Sour
Agrio
Hapan
Sprout-like
(similar a) Brote de una
Itumainen
planta
Stale
Pan rancio
Tunkkainen
Toasty
Tostado
Paahteinen
Waxy
(similar a) cera
Vahamainen
24 | Cátedra Villarreal | V. 4 | No. 1 | enero -junio | 2016 |