Foods 2020, 9, 619; doi:10.3390/foods9050619 www.mdpi.com/journal/foods
Evaluation of the Physicochemical and Sensory
Characteristics of Different Fig Cultivars for
the Fresh Fruit Market
, Alberto Martín
*, Margarita López-Corrales
, María de Guía Córdoba
Ana Isabel Galván
and Manuel Joaquín Serradilla
Food Science and Nutrition, School of Agricultural Engineering, University of Extremadura, Avda. Adolfo
Suárez s/n, 06071 Badajoz, Spain; email@example.com (C.P.); firstname.lastname@example.org (M.d.G.C.)
University Research Institute of Agro-Food Resources (INURA), Avda. de la Investigación s/n, Campus
Universitario, 06006 Badajoz, Spain
Department of Horticulture, Research Centre Finca La Orden-Valdesequera (CICYTEX), Junta de
Extremadura, Autovía Madrid-Lisboa s/n, 06187 Badajoz, Spain; email@example.com (M.L.-C.);
Department of Plant Sciences, Agrifood Technology Institute of Extremadura (INTAEX-CICYTEX), Junta de
Extremadura, Avda. Adolfo Suárez s/n, 06007 Badajoz, Spain; firstname.lastname@example.org
* Correspondence: email@example.com; Tel.: +34 -924-289-300 (ext. 86255)
Received: 6 April 2020; Accepted: 9 May 2020; Published: 12 May 2020
Abstract: Physicochemical and sensory properties of nine fig cultivars: ‘San Antonio’ (SA), ‘Blanca
Bétera’ (BB), ‘Brown Turkey’ (BT), ‘Tres Voltas L’Any’ (TV), ‘Banane’ (BN), ‘Cuello Dama Blanco’
(CDB), ‘Cuello Dama Negro’ (CDN), ‘Colar Elche’ (CE), and ‘De Rey’ (DR), were compared at three
different ripening stages. Weight, size, titratable acidity, pH, skin and flesh colours, firmness,
maturation index (MI), and volatile compounds were determined in samples from two consecutive
seasons, in addition to both descriptive and hedonic sensory analysis. The mechanical behaviour of
figs determined by firmness analysis and colour changes in both skin and flesh was the most
important trait for the discrimination of ripening stages. Notable differences among cultivars were
found in most of the parameters studied, in particular the inter-cultivar differences highlighted for
MI, pH, acidity, and skin colour analyses, followed by volatile compounds. Principal component
analysis (PCA) indicated that MI, pH, colour parameters of flesh (h and L*), and terpenes were the
best physicochemical indices to determine overall acceptability which is highly correlated with the
sensory attributes flesh colour and fruit flavour. The results suggested that CDN and SA showed
huge consumer acceptability among the fig cultivars studied.
Keywords: Ficus carica L.; sensory properties; volatile compounds; colour; firmness
The fig (Ficus carica L.) constitutes a spice that is widely grown in the Mediterranean area,
where the fig tree population has been present since its domestication . In this area, both fresh and
dried figs are an important part of the diet, being especially rich in nutrients such as sugar, fibre,
proteins, and minerals, but also in organic acids and polyphenols. Spain is the major producer of figs
in Europe, with approximately 36,380 tonnes, i.e., 38% of European production and 3% of world
production . Although most commercial production is of dried fruit, figs are also widely
consumed as fresh fruit.
Fresh fruit quality is determined by nutritional and bioactive composition, but also by other
parameters related to the sensory characteristics, including firmness, visual appearance, taste, and
aroma . Firmness is one of the primary attributes determining consumer acceptance, flesh
Foods 2020, 9, 619 2 of 16
firmness being the parameter used to determine the harvest time as well as the maturity grade
during postharvest of perishable fruit such as fig . Sugars, acids, and phenolic compounds
contribute to the taste and colour of figs, but also to the characteristic of flavour, which is dependent
mainly on the proper balance of the volatile chemical constituents [5,6]. On the other hand, the
aromatic compound profile of each variety is considered to be unique and has a great influence on
organoleptic characteristics and therefore on consumer acceptance [3,5]. Recently, the use of
techniques as solid-phase microextraction of headspace (HS-SPME) with gas chromatography
analysis with a mass detector (GC-MS) has made it possible to identify and quantify individually
this complex mixture of aromatic compounds, which mainly includes compounds such as alcohols,
aldehydes, ketones, esters and terpenoids. Additionally, this complex mixture depends on several
factors such as soil, climate, genotype, ripeness, and technological aspects . Compounds such as
ethyl acetate, hexanal, β-caryophyllene, limonene, (E)-2-hexenal, and octanal have been attributed
primarily to the aroma of fresh figs [8–10]. In addition, compounds such as furfural, benzaldehyde,
phenol, among others, have also shown a remarkable influence on the aroma of fresh fig [11,12]. The
volatile composition in fresh figs varies during the ripening process and with it the perception of
their sensory characteristics [5,11]. Thus, in fruit such as kiwifruit, a considerable increase in the
ester content and a decrease in the concentration of aldehydes throughout ripening has been
described . However, in the case of figs, the concentration of both compounds has also been
clearly influenced by genotype .
The majority of studies related to the characterisation of quality parameters of fresh fig cultivars
approach specific quality aspects such as physicochemical properties [14–16], firmness, or aroma
compounds [5,10,11]. Regarding the volatile aroma profile of figs, some studies have focused on the
analysis of aromatic compounds from leaves, spirits, extracts and others [5,8–11]. So far, however,
only a few studies showed an overall view of the relation among physicochemical parameters and
sensory attributes taking into account different ripening stages. Thus, Crisosto et al.  reported the
influence of two ripening stages on different physicochemical traits as well as other parameters such
as the total antioxidant capacity of four fig cultivars currently grown in California, highlighting the
great impact of total soluble solids content on consumer acceptance. King et al.  also
characterised the sensory properties of 12 California-grown fresh fig cultivars from six different
sources, finding significant correlations between sensory and physicochemical data. These authors
highlighted the importance of selecting cultivars with strong flavours that remain firm as they
To our knowledge, a comprehensive and interannual study of both physicochemical and
sensory quality characteristics in several commercial cultivars of fresh fig has never been carried out.
In this research, we established the relation among these physicochemical and sensory parameters
and an overview of quality traits of the fig cultivars studied for fresh fruit market, taking into
account the changes associated with the different commercial ripening stages.
2. Materials and Methods
2.1. Plant Material and Experimental Design
The nine fig tree cultivars studied for fresh consumption in the order of ripening (early, middle
or late) were ‘San Antonio’(SA), ‘Blanca Bétera’ (BB), ‘Brown Turkey’(BT), ‘Tres Voltas L’Any’(TV)
as early cultivars, ‘Banane’(BN) as mid cultivar, ‘Cuello Dama Blanco’ (CDB), ‘Cuello Dama
Negro’(CDN), ‘Colar Elche’(CE), and ‘De Rey’ (DR) as late cultivars. Studies of morphological and
molecular characterisation conducted in Spanish germplasm fig indicate that ‘Cuello Dama Blanco’
and ‘Colar Elche’ are the same varieties as ‘Kadota’ and ‘Mission’, respectively [19–21]. All cultivars
were selected among those available in the national germplasm bank of the fig tree is located in the
research centre “Finca La Orden-CICYTEX” at an altitude of 223 m above sea level (latitude
38°85′19″ N, longitude −6°68′28″ W, Guadajira, Badajoz, Spain) based on parameters of fruit quality.
Figs were harvested manually from July to October. With respect to the experimental design, fig
samples were hand-collected when they were fully mature and were harvested during two
Foods 2020, 9, 619 3 of 16
consecutive seasons from an experimental orchard established in 2007 and previously described in
our previous work .
For the maturation study, three different ripening stages of each cultivar were selected
according to their skin colour and firmness as per expert harvester criteria. Stage 1 corresponded to
the greener fruit, whereas Stage 3 corresponded to mature fruit. For each physicochemical
determination, three replicates of 10 fruit for each ripening stage and cultivar were established per
year. All analyses were conducted using fresh fruit, but for analysis of volatile aroma profile, the
samples were weighed in vials and stored at −80 °C for later analysis.
2.2. Weight and Size
Both parameters were determined using an AE-166 balance (Mettler, Madrid, Spain) for weight
(g) and a DL-10 digital micrometre (Mitutoyo, Kawasaki, Japan) for size (mm).
2.3. Total Soluble Solids (TSS), Titratable Acidity (TA), pH, and Maturation Index (MI)
A model RM40 Mettler Toledo digital refractometer at 20 °C was used to measure total soluble
solids (TSS) in °Brix. On the other hand, 5 g aliquots of fig homogenate diluted to 50 mL with
deionised water from a Milli-Q water purification system (Millipore, Bedford, MA, USA) were used
to determine titratable acidity and pH, using a T50 Compact Stirrer for automatic titration (Mettler
Toledo, Madrid, Spain), titrating up to pH 7.8 with 0.1 mol L−1 NaOH and expressing the results in g
citric acid 100 g−1 fresh weight (FW).
The maturation index (MI) was calculated as described by Pereira et al. .
Skin and flesh colour of figs was measured according to Pereira et al. . The parameters
brightness (L*), chroma (C*) and hue angle (h*) were measured using a Konica Minolta CM600
spectrophotometer in accordance with the CIELab system.
A 6% deformation was applied by a 70 mm aluminium plate coupled to a TA.XT2i Texture
Analyser (Stable Micro Systems, Godalming, UK) to measure firmness in N mm−1 .
2.6. Determination of Volatile Compounds
The volatile profile from each ripening stage and cultivar was analysed by solid-phase
microextraction (SPME) with a 10 mm-long, 75 µm-thick fibre coated with
Carboxen™/polydimethylsiloxane (Supelco, Bellefonte, PA) as described by Serradilla et al. .
The volatile compounds were identified and semi-quantified using an Agilent 6890 GC/5973
MS system (Agilent Technologies) using a DB-5 (Agilent Technologies J&W, Santa Clara, CA, USA)
bonded fused silica capillary column, coated with 5% phenyl/95% polydimethylsiloxane (30 m × 0.32
mm inner diameter, 1.05 µm film thickness). For the identification of volatile profile, in addition to
using the NIST/EPA/NIH mass spectrum library (comparison quality > 90%), and Kovats indices,
which were calculated using a mixture of n-alkanes (R-8769, Sigma Chemical Co., St. Louis, MO,
USA) run under the same conditions , pure compounds under the same chromatographic
conditions were also used to confirm the identifications.
2.7. Sensory Analysis
For the sensory analysis, only samples from ripening Stages 2 and 3 were used. A trained panel
of 10 panellists, 6 women and 4 men between the ages of 30 and 50 years old, was used to carry out a
descriptive sensory analysis, assessing the parameters previously described in our previous work
. A numbered scale from 1 to 10 points was used. Each panellist evaluated a total of two
defect-free fruit per ripening stage and cultivar immediately after harvest and after reaching a flesh
Foods 2020, 9, 619 4 of 16
temperature of 20 °C under white lighting, airflow, and temperature (20–22 °C) controlled
conditions. Samples were presented to each panellist on plastic plates in random order using a
3-digit code for each sample, assessing the following sensory attributes: external appearance, skin
colour, flesh colour, firmness, sweetness, acid, bitter, juiciness, presence of seeds, and fruit flavour.
One session was conducted per week throughout the harvest period. Additionally, hedonic tests,
using a numbered scale from 1 to 10 points, were performed to evaluate overall acceptability with a
total of 65 untrained consumers, but regular fig consumers, consisting of 35 women and 20 men aged
20 to 50 years old, was tested every 15 days throughout the harvest period.
2.8. Statistical Analysis
SPSS for Windows, 19.0 (SPSS Inc. Chicago, IL, USA) was used to carry out an analysis of
variance (ANOVA) of the mean values of physicochemical parameters, the area of volatile
compounds, and sensory characteristics. Tukey’s honestly significant differences (HSD) test (p ≤
0.05) was applied to separate means. Principal component analysis (PCA) was used to analyse the
relationships among the parameters studied, using ‘ripening stage’ and ‘cultivar’ as classification
3. Results and Discussion
3.1. Physicochemical Properties
Fig cultivar weights and sizes at the three ripening stages are shown in Figure 1A,B. The mean
weight of the cultivars studied was 46.5 g, with BT and BN showing the highest values for this
parameter. These same cultivars, along with CE and BB, showed a significant weight increase with
the ripening process, mainly between Stage 1 and Stage 2. For BT, mean weight increased from 54.8 g
(Stage 1) to 77.5 g (Stage 3). A similar weight increase associated with maturity stage was also found
for this cultivar by Crisosto et al. . On the contrary, TV presented weights less than 30 g for the
three stages studied. The weight tendency in fig cultivars was also observed for size, with 43.2 mm
as the mean value, although in this case, differences among maturity stages were not found (Figure
Foods 2020, 9, 619 5 of 16
Figure 1. Physicochemical parameters of the fig cultivars at the three ripening stages studied. Weight
(A); size (B); pH (C); titratable acidity (TA) (D); total soluble solids (TSS) (E); maturation index (MI)
(TSS/TA) (F). SA, San Antonio; BB, Blanca Bétera; BT, Brown Turkey; TV, Tres Voltas L’Any; BN,
Banane; CDB, Cuello Dama Blanco; CDN, Cuello Dama Negro; CE, Colar Elche; DR, De Rey. Tukey
HSD, Tukey’s honestly significant differences; SSB, Statistical significance bar.
The pH values of the fig cultivars studied are given in Figure 1C. The mean pH value was 5.8,
ranging from 5.16 (BB; Stage 1) to 6.39 (SA; Stage 3), which are slightly higher than those described
by other authors [25,26]. There was an increase in the pH values with increasing maturity, which
resulted in a less acid product. This change associated with ripening time was more evident for
cultivars BN, TV, and BB with a decrease of TA of more than 0.6 g citric acid 100 g−1 FW from Stage 1
to Stage 3 (Figure 1D). The mean value of this parameter for the cultivars studied was 1.2 g citric acid
100 g−1 FW, ranging from 0.72 (Stage 3 of SA) to 2.14 g citric acid 100 g−1 FW (Stage 1 of CDN). This
TA ratio is in line with those obtained by Çalişkan and Polat  for fig genotypes grown in the
Eastern Mediterranean Region of Turkey as well as some Turkish cultivars. In general, the early SA
cultivar was characterised by exhibiting the highest pH values and the lowest TA values at the most
advanced stages of ripening.
The fig cultivars studied had an average °Brix of 20.4 and MI of 200.6, all of them showing an
evident increase for these parameters during the ripening process of fig fruit, with the highest values
at maturation Stage 3 (Figure 1E,F). The TSS/acid ratio is directly related to fruit taste and in the fig’s
aptitude for the drying process . The differences between Stage 1 and Stage 3 were more than 6
°Brix for late cultivars CDN and CE, whereas the variation for mid cultivars BN and CDB was less
than 3 °Brix. For MI, the more relevant differences between ripening stages were found for early and
mid cultivars SA, BB, and BN, in contrast to late cultivars DR and CDB, which showed no significant
differences. In general, the °Brix values found in our study were in agreement with other reports
[26,28,29], whereas the MI values were slightly higher, mainly due to the low acidity presented in
the studied cultivars.
DR CDB BT SA CDN BN CE TV BB
DR CDB BT SA CDN BN CE TV BB
Weigh t (g )
DR CDB BT SA CDN BN CE TV BB
Maturation index (MI)
Stage 1 Stage 2 Stage 3
DR CDB BT SA CDN BN CE TV BB
TA (g 100g
DR CDB BT SA CDN BN CE TV BB
DR CDB BT SA CDN BN CE TV BB
Stage 1 Stage 2 Stage 3
SSB (± 3.14)
SSB (± 1.27)
0.05 Tukey HSD
SSB (± 0.16)
SSB (± 0.24)
0.05 Tukey HSD
SSB (± 1.02)
0.05 Tukey HSD
SSB (± 42.72)
Foods 2020, 9, 619 6 of 16
3.2. Colour and Firmness
The colour parameters of the fig varieties studied are shown in Figure 2. There were evident
differences in the mean values of skin colour parameters between the dark (CDN and CE),
purple/yellow (SA, BT, and DR), and green cultivars (CDB, TV, BN, and BB), but also between
ripening stages. An increase of h* values was observed for CDN and CE at Stage 3, whereas for the
rest of the varieties the decrease of L* values was the more relevant change. Crisosto et al. 
described a significant increase of h* values at the higher maturity stage studied for cv. Mission (syn.
CDN and CE). Likewise, differences were also observed in flesh colour parameters among different
cultivars of figs (p > 0.05), mainly related to the coordinates L* and h* (Figure 2B). The cultivars CDB
and SA showed the highest values for both colour parameters mentioned, with L* values of 55.53
and 59.19 and h* values of 73.47 and 71.39, respectively. On the contrary, the lowest values of L* and
h* in fig flesh were found for CDN and BT (lower than 46.39 and 48.06 for L* and 46.28 and 46.91 for
h*, respectively). These results agree with those for some Turkish fig cultivars which showed great
variability among cultivars with mean values of 53.79 for L* and 42.37 for h* .
Figure 2. Colour parameters of the fig cultivars at the three ripening stages studied. Skin (A); flesh
(B). L*, Brightness; C*, Chroma and h*, Hue angle.
On the other hand, there were also differences in the flesh colour among the three ripening
stages of the fig cultivars studied. In this case, C* values decreased during the ripening process for
all fig cultivars studied as a consequence of flesh darkening due to the accumulation of
anthocyanins, this change being more evident in the dark-skinned cultivars CDN and BT. A negative
correlation between total anthocyanins in flesh fruit and the chromatic parameter chroma has been
Foods 2020, 9, 619 7 of 16
described for several fruit such as sweet cherry . After the visual appearance, firmness is the
most relevant factor that determines the acceptability of fleshy fruit such as figs , firmness being a
relevant component. Firmness values decreased during the maturation of figs, this process being
cultivar-dependent (Figure 3). Initially, the mean fig firmness values were 2.57 N mm
at Stage 1,
decreasing to 0.75 N mm
at Stage 3. Our results during the ripening process of fig cultivars were
similar to those obtained by other authors [17,22]. Regarding cultivar differences, the firmness
values of SA and BB were higher than those of most of the cultivars studied, whereas TV presented
firmness values significantly lower than the rest of the cultivars for all ripening stages studied.
Figure 3. Firmness values of the fig cultivars at the three ripening stages studied.
3.3. Volatile Compounds
A total of 68 compounds were identified in the fig cultivars studied using HS/SPME and
GC/MS (Table 1). These volatile compounds were classified as aldehydes (20), hydrocarbons (9),
furans (8), alcohols (4), terpene compounds (4), ketones (3), acids (3), esters (3), pyranone derivates
(2), pyrimidines (1), and ethers (1). Similar findings in fresh figs were described in previous studies
Among the main volatile compounds that define fresh figs’ aroma profile are aldehydes .
These compounds represent a mean percentage of 18.25% of the total area, including linear,
branched, and aromatic aldehydes (Table 1). Regarding aromatic aldehydes, benzaldehyde (AL12)
accounted for 7.13% of the total area as mean value for the fig cultivars studied, being the most
abundant aldehyde found. They originate from the shikimic acid pathway and contribute greatly to
the characteristic aroma of fresh figs . The main linear aldehydes identified were (E)-2-hexenal
(AL9), nonanal (AL18), and hexanal (AL8) (2.65%, 2.41%, and 1.76% of the total area, respectively),
which have also been reported to be key to the volatile aroma profile of some fig cultivars [12,31].
These compounds exceeded 5% of the total area for some cultivars studied and are characterised by
exhibiting green leaf notes . Finally, among the branched aldehydes, 2-methyl-2-butenal (AL5)
showed a high percentage in most of the cultivars studied (1.09% of the total area), while the relative
concentration of the other branched aldehydes was less than 1% of the total area.
Foods 2020, 9, 619 8 of 16
Table 1. Volatile compounds in the fresh fig samples studied.
Volatile compounds CD† KI‡ ID§ RT¶ AAU| Percentage of area (%)#
Mean SD Min Max
3.20 0.32 15.15
Pentane, 2-methyl- H1 570 A 7.0 2 0.02
0.10 0.00 0.52
Pentane, 3-methyl- H2 585 A 7.7 6 0.05
0.20 0.00 1.01
Hexane¤ H3 600 A 8.9 241 1.84
1.64 0.23 7.29
Cyclopentane, methyl- H4 635 B 10.3 4 0.03 0.13 0.00 0.67
Heptane H5 700 A 14.7 10 0.08
0.18 0.00 0.55
Toluene H6 779 B 18.8 106 0.66
1.32 0.00 5.57
Ethylbenzene H7 864 B 23.6 17 0.10
0.21 0.00 0.86
p-Xylene H8 869 B 23.8 52 0.29
0.64 0.00 2.39
Tetradecane H9 1400 A 39.4 14 0.10 0.27 0.00 1.16
1.18 0.63 5.71
3-Buten-1-ol, 3-methyl- OL1 726 B 16.7 15 0.11 0.13 0.00 0.53
3-Heptanol OL2 894 B 24.7 44 0.40
0.71 0.00 2.66
Branched alcohol OL3 1028 D 29.8 26 0.30 0.49 0.00 1.90
Aromatic alcohol OL4 1080 D 31.7 292 1.69 0.81 0.40 3.40
C 5.2 90 0.60
0.42 0.00 1.95
2-Butenal, (E)- AL2 640 B 11.6 5 0.03 0.09 0.00 0.39
Butanal, 3-methyl- AL3 645 A 11.9 6 0.04 0.10 0.00 0.39
Butanal, 2-methyl- AL4 660 A 12.5 83 0.42 0.64 0.00 2.51
2-Butenal, 2-methyl- AL5 745 B 17.5 98 0.94 1.09 0.00 3.59
2-Pentanal, (E)- AL6 750 B 18.0 27 0.24 0.45 0.00 1.89
2-Butenal, 3-methyl- AL7 783 B 19.4 53 0.37 0.22 0.00 0.72
Hexanal AL8 800 A 20.2 221 1.76
1.44 0.22 5.98
2-Hexenal, (E)- AL9 853 A 22.8 305 2.65 2.97 0.50 15.15
Heptanal AL10 902 A 25.1 58 0.28
0.40 0.00 1.72
2,4-Hexadienal (E,E)- AL11 910 B 25.4 1 0.02 0.07 0.00 0.37
2-Heptenal AL12 952 B 27.4 4 0.04
0.06 0.00 0.23
Benzaldehyde AL13 956 A 27.8 985 7.13
6.33 1.45 32.44
Octanal AL14 1004 A 29.1 50 0.35 0.20 0.00 0.76
2,4-Heptadienal AL15 1010 B 29.4 14 0.11 0.24 0.00 1.09
Benzeneacetaldehyde AL16 1051 B 30.7 20 0.14 0.38 0.00 1.89
2-Octenal, (E)- AL17 1062 C 31.1 16 0.14 0.22 0.00 0.91
Nonanal AL18 1106 A 32.6 294 2.41 1.74 0.00 8.22
1164 B 34.5 7 0.04
0.07 0.00 0.27
Decanal AL20 1204 A 35.9 70 0.54 0.60 0.00 2.67
1.70 0.71 6.09
3-Heptanone K1 889 B 24.3 373 2.52
1.41 0.67 6.09
1,3-Cyclopentanone K2 992 C 28.8 106 0.64
0.57 0 1.89
2-Cyclopenten-1-one, 2-hydroxy-3-methyl- (Corylon) K3 1029 C 30.0 15 0.08 0.10 0 0.34
1.77 0.02 7.28
Acetic acid AC1
8.3 10 0.05
0.12 0 0.51
Hexanoic acid, 2-ethyl- AC2 1123 32.8 520 2.66 1.62 0 6.15
Nonanoic acid AC3 1277 37.3 64 0.41
0.35 0 1.13
20.31 0.69 65.61
Acetic acid, methyl ester ES1 554 A 6.2 42 0.38 0.30 0.01 1.15
Acetic acid, ethyl ester ES2 628 A 9.6 3496 26.53 20.11 0.67 65.26
Butanoic acid, methyl ester ES3 723 B 16.4 7 0.02 0.10 0 0.52
13.72 1.83 57.33
Furfural F1 830 A 21.9 966 4.24
2.32 0.51 8.21
2-Furanmethanol F2 859 B 23.0 903 4.39
2.06 0.97 8.2
2(3H)-Furanone, 5-methyl- F3 930 B 26.0 336 1.69 0.83 0.1 2.88
Furaneol F4 1058 B 30.8 49 0.30 0.60 0 2.31
Unknown furan 1 F5 1098 D 32.2 156 0.45 0.64 0 2.31
5-(Hydroxymethyl)-2(5H)-furanone F6 1178 B 34.9 65 0.33 0.57 0 2.34
Unknown furan 2 F7 1197 D 35.5 509 2.14 1.34 0 4.8
5-Hydroxymethylfurfural F8 1224 B 36.5 2847 9.52 8.41 0 33.38
7.15 0.81 26.64
2H-Pyran, 3,4-dihydro- P1
25.5 313 1.59 0.76 0.4 2.97
3-Hydroxy-2,3-dihydromaltol P2 1140 B 34.2 2727 11.06
6.54 0.36 25.06
α-Pinene T1 940 B 26.7 9 0.05
0.14 0 0.65
Unknown monoterpene T2 1006 D 29.2 9 0.04 0.11 0 0.39
Limonene T3 1030 A 30.3 38 0.19 0.29 0 1.18
Linalool T4 1098 B 32.6 101 0.89 1.41 0 4.35
Ethyl ether ET1 A 5.3 10 0.07
0.19 0 0.91
Thymine M1 1075 B 31.6 855 5.85 3.28 0.58 11.5
† CD: compound code used. ‡ KI: Kovats retention index. § ID: reliability of identification: A,
identified by a comparison to standard compounds; B, tentatively identified by the NIST/EPA/NIH
Foods 2020, 9, 619 9 of 16
mass spectrum library (comparison quality >90%) and Kovats retention index; C, tentatively
identified by the NIST/EPA/NIH mass spectrum library (comparison quality >90%); D, tentatively
identified by the NIST/EPA/NIH mass spectrum library (comparison quality <90%). ¶ RT: retention
time. |AAU: arbitrary area units. # %: relative abundance. ¤ In bold: volatile compound included in
PCA analysis, according to high relevance in the volatile profile of the fig cultivar studied.
Hexane (H3) was the most abundant hydrocarbon, a chemical class that involved 3.18% of the
total area as mean value for these studied cultivars, including branched and aromatic compounds
(Table 1). In addition, short-chain alkanes have also been reported, although in lower concentration,
to be present in fig fruit , described as non-contributors to fruit flavour. On the contrary,
particularly in light and yellow-green cultivars, other relevant compounds in the aroma profile of
figs are alcohols (2.50% of the total area), whether linear, branched, or aromatic. (Table 1) .
3-Heptanol (OL2) is associated with fresh green odours and green leaf notes that are typically linked
to fruit such as yellow passionfruit .
Furans represented 23.05% as a mean percentage, ranging significantly between 1.83% and
57.33% of the total area, mainly due to the variability found for 5-hydroxymethylfurfural in the
cultivars studied (Table 1). Both 5-Hydroxymethylfurfural and furfural have been associated with
sweet flavour notes and indeed both have been identified among the characteristic aroma
compounds of several other fruit, such as kiwifruit . Additionally, furans and the derivates of
pyranones have been reported to be derived directly from carbohydrates . Pyranones were the
fourth most abundant chemical class in the cultivars studied, representing a mean percentage of
12.65% of the total area. 3-Hydroxy-2,3-dihydromaltol (P1) was characterised by being the dominant
compound within this family with 11.06% of the total area for these studied cultivars, describing its
aromatic note as caramel. Furan and pyranone derivates show an outstandingly low odour
threshold and are also considered to be primary contributors to the volatile profile of dried figs
Regarding ketones (3.24% of the total area), a total of three of these compounds were identified
(Table 1). 3-Heptanone was the ketone detected at highest concentrations in the cultivars studied.
Pino et al.  considered this compound as a green fatty aroma note, and it has been described
among the most important aroma-active volatiles of fruit such as choch (Lucuma hypoglauca
With respect to acids (3.12% of the total area), acetic acid (AC1), nonanoic acid (AC3), and the
most abundant 2-ethyl hexanoic acid (AC2) were detected (Table 1). This last compound has a
negative effect on the overall aroma of fruit derivates, possessing an unpleasant odour with slightly
putrid notes . 2-Ethylhexanoic acid has been also reported as a regular food packaging material
contaminant . In our study, cultivars showed a mean percentage of 2.66% of the total area for this
Esters, with 26.93% of the total area, represented the other main group of aromatic compounds
detected in these fruit. These compounds are generated from the esterification of alcohols and
acyl-CoA derivates, highlighting the concentration shown by ethyl acetate (ES2), at 26.53% of the
total area, as the mean value in the cultivars studied (Table 1). This result could suggest that this
volatile compound may be relevant to the aroma profile of these cultivars. This compound was also
described in two Portuguese fig varieties (‘Branca Tradicional’ and ‘Pingo do Mel’), although the
concentration shown by these cultivars was lower than that obtained in this study .
Other compounds, which represented around 1% of the total area of volatile compounds of the
cultivars studied, were monoterpenes such as α-pinene (T1), limonene (T3), and linalool (T4) (Table
1), which are among the most frequent groups of aroma compounds identified in figs [5,8,11,33].
In general, remarkable fluctuations were observed in the volatile profiles of the samples studied
according to the values of standard deviations and ranges shown in Table 1. To understand the role
of both factors, cultivar and ripening stage, in this variability, a PCA was performed with the major
volatile compounds (>0.85% of total area). The PCA showed clear differences in the volatile profile
among the cultivars, and a limited influence of the ripening stages selected in this work (Fig. 4 and
5). Thus, the cultivars SA, BT, and BN showed a lower amount of the main volatile compounds
Foods 2020, 9, 619 10 of 16
compared to DR, BB, and CDB. Concretely, the cultivar DR was associated with high concentrations
of the main furans (F1, F2, F3, F7, and F8), pyranones (P1 and P2), and, to a lesser extent, aldehydes
such as benzaldehyde (AL13) and hexanal (AL8). In the case of cultivars BB and CDB, their volatile
profiles are highlighted for the high amount of 2-methyl-2-butenal (AL5), decanal (AL20), linalool
(T4), and hexane (H3) among other compounds (Figures 4 and 5). The differences found in both
physicochemical properties and volatile profiles of the fig cultivars studied may have a relevant
impact on their sensory parameters.
Figure 4. Loading plot (A) and score plot (B) after principal component analysis of the varieties,
ripening stages, and volatile compounds in the plane by two first principal components (PC1 and
PC2). Code letters for volatile compounds are shown in Table 1.
3.4. Sensory Analysis
-1 -0.5 0 0.5 1
PC 2 (14.0%)
PC 1 (49.38%)
-1 -0.5 0 0.5 1 1.5 2
PC 2 (14.0%)
PC 1 (49.38%)
0.6 0.7 0.8 0.9 1
Foods 2020, 9, 619 11 of 16
ANOVA of sensory descriptors (Table 2) shows significant differences among most of the fig
cultivars studied but not between ripening Stages 2 and 3, with the exception of external appearance
which was better for samples of ripening Stage 2. CDN and CE cultivars obtained the best scores
(higher than 7.00) for the descriptive parameters—external appearance, skin colour, and
texture—whereas BB was scored highest by panellists for sweetness (5.56), although no significant
differences were found for this parameter. The highest scores for fresh colour and fruit flavour
descriptors were found in CDB and SA samples (5.97–6.70 and 5.67–6.25, respectively), both
parameters showing a high degree of correlation with acceptability in the hedonic test. In fact, these
cultivars showed the best scores for the overall acceptability descriptor (6.71 and 6.65). In the case of
SA, the high score for the descriptor fruit flavour was not correlated with the relatively low number
of volatile compounds found for this cultivar (Figures 4 and 5), showing that the flavour is a
complex combination of not only olfactory but also gustatory-tactile and kinaesthetic sensations .
On the contrary, BT presented the worst score for acceptability (4.37), which was clearly related to its
lowest scores for skin colour (5.04), fresh colour (4.02), and fruit flavour (4.05). The difference
between these results and those found by other authors for BT can be partially explained by the
better sensory quality of pollinated fruit with respect to the parthenocarpic fruit used for our study
3.5. Interactions between the Analytical Parameters and Sensory Characteristics
PCA was carried out for the whole set of data to obtain an interpretable overview of the main
information. Samples at ripening Stage 1 were excluded as they were not sensorially analysed.
Figure 5 shows the two-way loadings and score plots, where PC2 and PC3 were plotted against PC1
to show a high percentage of the total variance (48.20%–41.04%). High values for MI, pH, skin C*,
and skin L* were explained by the positive axis of PC1 and were related to SA and, to a lesser extent,
to CDB and BB. These parameters showed a negative correlation with acidity, skin h, and the
sensory descriptors acid taste, skin colour, external appearance, and texture, which were highlighted
in CDN and CE. The second PC was mainly explained by most of the chemical families of the
volatile compounds located in the extreme of the positive axis, relating high values of those to the
cultivar DR. On the contrary, the cultivars with the highest weight and calibre values, BT and BN,
were associated with the negative axis of PC2 and therefore with a poor concentration of volatile
compounds. Variability of the sensory descriptors sweetness and juiciness was mainly explained by
the positive axis of PC3, showing a clear negative correlation with the descriptor bitter taste, high
values being those associated with BB.
Regarding acceptability, high scores in the hedonic test were correlated with MI, pH, flesh h,
and terpenes, in addition to the above-mentioned sensory descriptors (flesh colour and fruit
flavour). Likewise, the association of high acceptability scores with CDB (plots defined by PC1 and
PC2) was again observed, but also with SA (plots defined by PC1 and PC3).
In conclusion, the characteristics of the nine fig cultivars included in this study can be explained
on the basis of the physicochemical and sensory properties studied, most of them showing notable
differences among them. This confirms the relevance of characterising fig cultivars for the
assessment of potential consumer acceptability. In agreement with the PCA results, MI, pH, acidity,
and skin colour show the highest variability among the parameters studied in fresh figs, followed by
volatile compounds. Acceptance of the cultivars studied is associated with high values for MI, pH,
flesh colour parameters (h and L*), and terpenes, highlighting the sensory attributes flesh colour and
fruit flavour. In particular, CDN and SA showed high consumer acceptability among the fig
Foods 2020, 9, 619 12 of 16
Table 2. Scores of the sensory descriptive attributes and overall acceptability for the samples of the different cultivars studied and for ripening Stages 2 and 3.
External Appearance Skin Colour Flesh Colour Firmness Sweetness Acid Bitter Juiciness Seeds Fruit Flavour Overall Acceptability
DR 6.30 5.42 5.58 6.35 3.76 1.30 2.93 5.18 3.33 5.84 6.11
CDB 6.53 6.18 5.97 6.80 3.11 1.41 5.14 4.71 4.93 6.70 6.71
BT 6.16 5.04 4.02 5.82 3.97 1.61 4.10 5.00 4.48 4.05 4.37
SA 5.62 5.85 5.67 5.54 4.09 1.22 3.67 5.46 3.81 6.25 6.65
CDN 7.12† 7.09 5.23 7.24 3.38 2.32 3.95 4.39 2.64 4.35 5.01
BN 4.92 6.43 5.22 4.97 3.46 2.17 3.80 5.49 3.63 4.83 5.08
CE 6.88 6.27 5.09 6.77 3.25 2.11 3.98
4.74 3.67 4.81 5.68
TV 4.97‡ 6.24 5.71 5.19 2.86 1.61 3.69 4.39 3.32 5.91 4.90
BB 5.59 5.15 5.03 5.26 4.28 1.84 3.00 5.56 4.88 5.21 5.57
2 6.24 6.09 5.23 6.19 3.37 1.69 3.61 4.95 3.77 5.10 5.40
3 5.77 5.84 5.33 5.80 3.77 1.77 4.00 5.03 3.93 5.55 5.73
PC§ 0.000 0.000 0.006 0.000 0.389 0.032 0.008 0.130 0.000 0.000 0.000
Tukey CI¶ ± 1.60 ±1.49 ±1.64 ±1.52 ±2.26 ±1.10 ±1.89 ±1.75 ±1.77 ±1.73 ±1.45
PS# 0.037 0.230 0.684 0.075 0.210 0.651 0.143 0.738 0.526 0.064 0.117
† Numbers in bold type indicate the maximum score of the attribute. ‡ Underlined numbers in italics indicate the minimum score of the attribute. § Pc: p-value for
cultivar factor. ¶ CI: confidence interval for post-hoc Tukey HSD test. # Ps: p-value for ripening stage factor.
Foods 2020, 9, 619 13 of 16
Figure 5. Loading plots (A and C) and score plots (B and D) after principal component analysis of the varieties, ripening stages, physicochemical, and sensory
parameters and in the planes by three first principal components (PC1, PC2, and PC3). Physicochemical parameters (△), weight (W), calibre (C), °Brix (°B), acidity
(Ac), pH, maturation index (IM), ﬁrmness (Firm). Colour parameters (□), skin h* (hsk), C* (C*sk), L* (L*sk); fesh h* (hfl), C* (C*fl), L* (L*fl). Chemical families of
volatile compounds (ᚕ), aldehydes (AL), acids (Ac), alcohols (OL), ketones (K), furans (F), pyrazines (P), hydrocarbons (H), terpenes (T), esters (ES). Sensory
parameters (○), external appearance (Ext), skin colour (Csk), fresh colour (CFl), texture (Tex), sweetness (Swe), bitterness (Bit), juiciness (Jui), fruit flavour (Fru),
overall acceptability (Acep).
-2.5 -2 -1.5 -1 -0.5 0 0 .5 1 1.5
PC 3 (13.4%)
PC 1 (27.7%)
Csk CFl Fru
-1 -0.5 0 0.5 1
PC 3 (13.4%)
PC 1 (27.7%)
-1 -0.5 0 0.5 1
PC 1 (27.7%)
-3 -2 -1 0 1 2
PC 2 (20.5%)
PC 1 (27.7%)
Foods 2020, 9, 619 14 of 16
Author Contributions: Conceptualisation, C.P.; methodology, M.L.-C.; software, A.I.G.; validation, A.M.;
formal analysis, C.P., A.M.; investigation, C.P., A.I.G., A.M.; resources, A.M., M.J.S.; data curation, C.P.;
writing—original draft preparation, A.M.; writing—review and editing, A.M., M.L.-C., M.J.S., M.d.G.C.;
visualisation, A.M.; supervision, project administration, M.L.-C., M.d.G.C.; funding acquisition, M.L.-C. All
authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria
(INIA) Ministerio de Ciencia e Innovación (Spain), grant number RTA2010-00123; and Junta of Extremadura,
grant nunmbers GR18165 and GR18196. Cristina Pereira is the beneficiary of the pre-doctoral grant (BOE n◦ 31,
Sec. III. pág.: 12862) from INIA.
Acknowledgments: The authors are grateful to M. Cabrero and J. Barneto for technical assistance.
Conflicts of Interest: The authors declare no conflict of interest.
1. Ercisli, S.; Tosun, M.; Karlidag, H.; Dzubur, A.; Hadziabulic, S.; Aliman, Y. Color and antioxidant
characteristics of some fresh fig (Ficus carica L.) genotypes from northeastern Turkey. Plant Food Hum.
Nutr. 2012, 67, 271–276.
2. FAOSTAT, FAO database. 2017. Available online: http://faostat.fao.org (accessed on 15 January 2020).
3. Valero, D.; Serrano, M. Postharvest Biology and Technology for Preserving Fruit Quality; CRC Press: Boca
Raton, FL, USA, 2010.
4. Contador, L.; Shinya, P.; Infante, R. Texture phenotyping in fresh fleshy fruit. Sci. Hortic. 2015, 193, 40–46.
5. Oliveira, A.P.; Silva, L.R.; de Pinho, P.G.; Gil-Izquierdo, A.; Valentão, P.; Silva, B.M.; Pereira, J.A.;
Andrade, P.B. Volatile profiling of Ficus carica varieties by HS-SPME and GC–IT-MS. Food Chem. 2010, 123,
6. Pereira, J.; Pereira, J.; Câmara, J.S. Effectiveness of different solid-phase microextraction fibres for
differentiation of selected Madeira island fruits based on their volatile metabolite profile—Identification of
novel compounds. Talanta 2011, 83, 899–906.
7. Forney, C.F.; Kalt, W.; Jordan, M.A. The composition of strawberry aroma is influenced by cultivar,
maturity, and storage. HortScience 2000, 35, 1022–1026.
8. Gibernau, M.; Buser, H.R.; Frey, J.E.; Hossaert-McKey, M. Volatile compounds from extracts of figs of
Ficus carica. Phytochem 1997, 46, 241–244.
9. Grison, L.; Edwards, A.A.; Hossaert-McKey, M. Interspecies variation in floral fragrances emitted by
tropical Ficus species. Phytochem 1999, 52, 1293–1299.
10. Grison-Pigé, L.; Hossaert-McKey, M.; Greeff, J.M.; Bessière, J.M. Fig volatile compounds a first
comparative study. Phytochem 2002, 61, 61–71.
11. Gozlekci, S.; Kafkas, E.; Ercisli, S. Volatile compounds determined by HS/GC-MS technique in peel and
pulp of fig (Ficus carica L.) cultivars grown in Mediterranean region of Turkey. Not. Bot. Horti Agrobot
Cluj-Napoca 2011, 39, 105–108.
12. Villalobos, M.C.; Serradilla, M.J.; Martín, A.; Aranda, E.; López-Corrales, M.; Córdoba, M.G. Influence of
modified atmosphere packaging (MAP) on aroma quality of figs (Ficus carica L.). Postharvest Biol. Technol.
2018, 136, 145–151.
13. Wang, M.Y.; MacRae, E.; Wohlers, M.; Marsh, K. Changes in volatile production and sensory quality of
kiwifruit during fruit maturation in Actinidia deliciosa ‘Hayward’ and A. chinensis ‘Hort16A’. Postharvest
Biol. Technol. 2011, 59, 16–24.
14. Polat, A.A.; Çalişkan, O. Fruit characteristics of table fig (Ficus carica L) cultivars in subtropical climate
conditions of the Mediterranean region. New Zeal J. Crop. Hort. 2008, 36, 107–115.
15. Çalişkan, O.; Polat, A.A. Effects of genotype and harvest year on phytochemical and fruit quality
properties of Turkish fig genotypes. Span. J. Agric Res. 2012, 10, 1048–1058.
16. Gozlekci, S. Pomological traits of fig (Ficus carica L.) genotypes collected in the west Mediterranean region
in Turkey. J. Anim. Plant Sci. 2011, 21, 646–652..
17. Crisosto, C.H.; Bremer, V.; Ferguson, L.; Crisosto, G.M. Fresh fig (Ficus carica L.) cultivars harvested at two
maturity stages. HortScience 2010, 45, 707–710.
18. King, E.S.; Hopfer, H.; Haug, M.T.; Orsi, J.D.; Heymann, H.; Crisosto, G.M.; Crisosto, C.H. Describing the
appearance and flavor profiles of fresh fig (Ficus carica L.) cultivars. J. Food Sci. 2012, 77, 419–429.
Foods 2020, 9, 619 15 of 16
19. Giraldo, E.; Lopez-Corrales, M.; Hormaza, J.I. Optimization of the management of an ex-situ germplasm
bank in common fig with SSRs. J. Am. Soc Hortic Sci. 2008, 133, 69–77.
20. Giraldo, E.; Viruel, M.A.; López-Corrales, M.; Hormaza, J.I. Characterisation and cross-species
transferability of microsatellites in the common fig (Ficus carica L.). J. Hortic. Sci. Biotechnol. 2005, 80, 217–
21. López-Corrales, M.; Gil, M.; Pérez, F.; Cortés, J.; Serradilla, M.J.; Chome, P.M. Variedades de Higuera,
Descripción y Registro de Variedades; Ministerio de Medio Ambiente y Medio Rural y Marino: Madrid, Span,
22. Pereira, C.; López-Corrales, M.; Martín, A.; Villalobos, M.C.; Córdoba, M.G.; Serradilla, M.J.
Physicochemical and nutritional characterization of brebas for fresh consumption from nine fig varieties
(Ficus carica L.) grown in Extremadura (Spain). J. Food Qual. 2017, 2017, 1–12.
23. Serradilla, M.J.; Martín, A.; Ruiz-Moyano, S.; Hernández, A.; López-Corrales, M.; Córdoba, M.G.
Physicochemical and sensory characterisation of four sweet cherry cultivars grown in Jerte Valley (Spain).
Food Chem. 2012, 133, 1551–1559.
24. Kondjoyan, N.; Berdagué, J.L. A Compilation of Relative Retention Indices for the Analysis of Aromatic
Compounds; Ed. du Laboratoire Flaveur, INRA de Theix, Saint-Gènes-Champanelle, France, 1996.
25. Silva, L.C.A.S.; Harder, M.N.C.; Arthur, P.B.; Lima, R.B.; Modolo, D.M.; Arthur, V. Physical-chemical
characteristics of figs (Ficus carica L.) preready to submitted to ionizing radiation. In Proceedings of the
International Nuclear Atlantic Conference, Rio de Janeiro,RJ, Brazil, 27 September 2009 to 2 October 2009.
26. Ersoy, N.; Gozlekci, S.; Gok, V.; Yilmaz, S. Fig (Ficus carica L.) fruit, some physical and chemical properties.
Acta Hortic. 2017, 1173, 329–334.
27. Çalişkan, O.; Polat, A.A. Fruit characteristics of fig cultivars and genotypes grown in Turkey. Sci. Hortic.
2008, 115, 360–367.
28. Mars, M.; Chebli, T.; Marrakchi, M. Multivariate analysis of fig (Ficus carica L.) germplasm in southern
Tunisia. Acta Hortic. 1998, 480, 75–78.
29. Khapre, A.P.; Satwadhar, P.N.; Syed, H.M. Studies on processing technology and cost estimation of fig
(Ficus carica L.) fruit powder enriched Burfi (Indian cookie). IASET 2015, 7, 621–624.
30. Gonçalves, B.; Silva, A.P.; Moutinho-Pereira, J.; Bacelar, E.; Rosa, E.; Meyer, A.S. Effect of ripeness and
postharvest storage on the evolution of color and anthocyanins in cherries (Prunus avium L.). Food Chem.
2007, 103, 976–984.
31. Oliveira, A.P.; Silva, L.R.; Andrade, P.B.; Valentão, P.; Silva, B.M.; Pereira, J.A.; de Pinho, P.G.
Determination of low molecular weight volatiles in Ficus carica using HS-SPME and GC/FID. Food Chem.
2010, 121, 1289–1295.
32. Li, J.; Tian, Y.; Sun, B.; Yang, D.; Chen, J.P.; Men, Q.M. Analysis on volatile constituents in leaves and fruits
of Ficus carica by GC-MS. Chin. Herb Med. 2012, 4, 63–69.
33. Trad, M.; Ginies, C.; Gaaliche, B.; Renard, C.M.; Mars, M. Does pollination affect aroma development in
ripened fig (Ficus carica L.) fruit. Sci. Hortic. 2012, 134, 93–99.
34. Matsui, K. Green leaf volatiles, hydroperoxide lyase pathway of oxylipin metabolism. Curr. Opin. Plant.
Biol. 2006, 9, 274–280.
35. Werkhoff, P.; Güntert, M.; Krammer, G.; Sommer, H.; Kaulen, J. Vacuum headspace method in aroma
research, flavor chemistry of yellow passion fruits. J. Agric. Food Chem. 1998, 46, 1076–1093.
36. Schwab, W.; Davidovich-Rikanati, R.; Lewinsohn, E. Biosynthesis of plant-derived flavor compounds.
Plant. J. 2008, 54, 712–732.
37. Mujić, I.; Kralj, M.B.; Jokić, S.; Jug, T.; Šubarić, D.; Vidović, S.; Jarni, K. Characterisation of volatiles in dried
white varieties figs (Ficus carica L.). Int. J. Food Sci. Technol. 2014, 51, 1837–1846.
38. Pino, J.; Moo-Huchin, V.; Sosa-Moguel, O.; Sauri-Duch, E.; Cuevas-Glory, L. Characterization of
aroma-active compounds in choch (Lucuma hypoglauca Standley) fruit. Int. J. Food Prop. 2017, 20, 444–448.
39. Siegmund, B.; Murkovic, M. Changes in chemical composition of pumpkin seeds during the roasting
process for production of pumpkin seed oil (part 2, volatile compounds). Food Chem. 2004, 84, 367–374.
40. Elss, S.; Grünewald, L.; Richling, E.; Schreier, P. Occurrence of 2-ethylhexanoic acid in foods packed in
glass jars. Food Addit. Contam. 2004, 21, 811–814.
41. Spence, C. Multi-sensory integration & the psychophysics of flavour perception. In Food Oral Processing
Fundamentals of Eating and Sensory Perception; Chen, J., Engelen, L., Eds.; Blackwell Publishing: Oxford, UK,
2012; pp. 203–219.
Foods 2020, 9, 619 16 of 16
42. Rosianski, Y.; Freiman, Z.E.; Cochavi, S.M.; Yablovitz, Z.; Kerem, Z.; Flaishman, M.A. Advanced analysis
of developmental and ripening characteristics of pollinated common-type fig (Ficus carica L.). Sci. Hortic.
2016, 198, 98–106.
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).