Food and Bioproducts Processing 1 1 7 ( 2 0 1 9 ) 321–328

Contents lists available at ScienceDirect

Food and Bioproducts Processing

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

Optimization of lycopene extraction from tomato

peels industrial by-product using maceration in
refined olive oil

Mouna Kehili a,d,∗ , Sami Sayadi b , Fakher Frikha c , Ayachi Zammel d ,

Noureddine Allouche e
a Laboratory of Environmental Bioprocesses, Centre of Biotechnology of Sfax, University of Sfax, PO Box 1177, 3018,
Sfax, Tunisia
b Center for Sustainable Development, College of Arts and Sciences, Qatar University, 2713, Doha, Qatar
c Faculty of Sciences of Sfax, University of Sfax, PO Box 1171, 3000, Sfax, Tunisia
d Ayachi Group Industry, El Mansoura, 6131, Siliana, Tunisia
e Laboratory of Organic Chemistry, LR17ES08 (Natural Substances Team), Faculty of Sciences of Sfax, University of

Sfax, PO Box 1171, 3000, Sfax, Tunisia

a r t i c l e i n f o a b s t r a c t

Article history: Lycopene, a nutraceutical compound, was extracted from a Tunisian industrial tomato
Received 4 May 2019 peels by-product using maceration in refined olive oil (ROO). To do, a Box–Behnken design
Received in revised form 22 July 2019 was applied to maximize the lycopene recovery within ROO using variable temperatures
Accepted 7 August 2019 (40–80 ◦ C), agitation speeds (200–400 rpm) and tomato peels to ROO ratios (2.5–5.5%, w/v).
Available online 13 August 2019 Results showed that 99.3% of the initial lycopene content, being of 1244.50 ± 30.41 mg/kg
on dry basis, was extracted using a biomass to oil ratio of 2.5% (w/v), a temperature of
Keywords: 80 ◦ C and a magnetic stirring of 400 rpm for 45 min, resulting in a concentration of 35 mg
Tomato peels residues lycopene/kg ROO. Importantly, this study presented a combined solution towards solving
Oil maceration the pollution problem connected with tomato processing while enriching ROO in bioactive
Lycopene extraction molecules, namely lycopene, preventing it from the potential drawbacks of lipid oxidation
Box–Behnken design and increasing its nutritional and health benefits.
Prediction model © 2019 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction
In the last decades, extensive research has been focused on the recov-
ery of antioxidant compounds consisting of natural and synthetic
molecules that are able to directly trap the free radicals and the reac-
tive oxygen species (ROS) in order to reduce their oxidative damage
to lipids, proteins, and deoxyribonucleic acid (DNA). Besides, antiox-
idants may act indirectly to up-regulate the endogenous antioxidant
defenses and inhibit the ROS production which has been claimed to
cause more than 100 diseases including cancer, atherosclerosis, neu-
rological degeneration, and cataracts (Farzaneh et al., 2018; Dolatabadi
et al., 2016; Khlebnikov et al., 2007). In this context, much research
has been extended on the valorization of agro-food products and their
industrial by-products for the recovery of natural and low-cost antiox-
idants and their use as alternatives to synthetic antioxidants, e.g.
butylated hydroxyanisole (BHA) and butylated hydroxytoluene (BHT)
(Dolatabadi et al., 2016). Effectively, the synthetic antioxidants have
been recently claimed to be liver-damaging and carcinogenic, there-
fore they have been restricted in different parts of the world (Moure
et al., 2001; Torbica et al., 2016; Dolatabadi et al., 2016). Importantly,
great interest has been given to the supplementation of natural antiox-
idants to refined vegetable oils as they are partially deprived from their
original antioxidants, e.g. the phenolic compounds, during the refin-
ing process which makes them much prone to the oxidative reactions
(García et al., 2006).

∗
Corresponding author at: Centre of Biotechnology of Sfax, University of Sfax, PO Box 1177, 3018, Sfax, Tunisia.
E-mail addresses: kehili.mouna@yahoo.fr (M. Kehili), sami.sayadi@gmail.com (S. Sayadi), frikha fakher@yahoo.fr (F. Frikha),
direction@ayachi-group.com (A. Zammel), noureddineallouche@yahoo.fr (N. Allouche).
https://doi.org/10.1016/j.fbp.2019.08.004
0960-3085/© 2019 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
322 Food and Bioproducts Processing 1 1 7 ( 2 0 1 9 ) 321–328
As highlighted in our study by Kehili et al. (2017a), in spite of their
well-known antioxidative properties, carotenoids including lycopene,
have attracted very limited attention as natural additives for the pre-
vention of vegetable oils from the oxidative reactions. In this regard,
tomatoes were deduced as the major source of dietary lycopene with
tomato peels containing five times more lycopene than tomato pulp
(Strati et al., 2015). Therewith, in the last decade, Tunisia was ranked
among the top ten countries, worldwide, for the industrial transforma-
tion of tomatoes resulting in high quantities of wastes ranging between
20,000 tons and 30,000 tons during the summer season with 56% are
tomato peels, on dry weight basis (Kehili et al., 2016). Although tomato
industrial by-products were recognized as a potential source of value-
added products, they are currently discharged in the nature or used as
animal feed (Papaioannou and Karabelas, 2012; Torbica et al., 2016).
On the other hand, Tunisia represents a very important country in
the olive oil producing world, being the largest African exporter and
the fourth worldwide olive oil exporter after Spain, Italy and Greece
with an annual production exceeding 170,000 tons (Issaoui et al., 2010;
Baccouri et al., 2008).
Interestingly, in our previous study by Kehili et al. (2017a), we
provided a new insight on the use of lycopene-rich oleoresin (TPO),
extracted from tomato peels industrial by-product, as a natural addi-
tive for the stabilization of refined olive (ROO) and sunflower (RSO)
oils as these oils are widely consumed among the Tunisian popula-
tion (Ben-Ali et al., 2014). It was deduced that the lycopene-rich TPO
extract exhibited an effective dietary stabilization of both ROO and
RSO instead of the synthetic preservative, BHT, against the oxidation
reactions during a long storage period.
Generally, lycopene extraction from natural resources was com-
monly carried out via conventional solvent extraction (CSE) and
supercritical (CO2 ) extraction (SC-CO2 ). CSE was deemed to consume
large quantities of solvents and a long extraction time with a rel-
atively poor efficiency (Bakhshabadi et al., 2017; Dolatabadi et al.,
2016), whereas SC-CO2 extraction which is considered as a safe and
simple process with a remarkable solvating capacity requires a high
energy input and an expensive equipment (Xu and Pan, 2013; Shan
et al., 2011; Galanakis, 2013). Furthermore, new extraction methods
assisted with enzyme, microwave, ultrasound and surfactant pretreat-
ments are being investigated for carotenoids recovery from tomato
(Amiri-Rigi and Abbasi, 2016; Lianfu and Zelong, 2008; Papaioannou and
Karabelas, 2012; Galanakis, 2013). Still, these alternative methods have
not yet been totally exempt from hazardous solvents and they have
limited carotenoids recovery yields as compared to SC-CO2 extraction
(Papaioannou et al., 2015; Kehili et al., 2017b).
Interestingly, as lycopene is a hydrophobic pigment, it tends to dis-
solve much better in non-polar solvents and fats. Therefore, in the
present study, we intended to extract lycopene from tomato peels by-
product using ROO as a safe, environmentally friendly and solvent
free extraction medium. In order to maximize the lycopene extraction
yield and enable a cost effective production of a lycopene-rich ROO, the
maceration process was optimized using a Box–Behnken design with
variable temperatures of 40–80 ◦ C, agitation speeds of 200–400 rpm and
tomato peels to ROO ratios of 2.5–5.5% (w/v). The optimum extraction
conditions were then performed at variable maceration times ranging
from 15 to 150 min. The lycopene extraction yields were determined
based on the HPLC quantification of this carotenoid. Intriguingly, this
study aimed not only to increase the tomato industrial bioeconomy
and solve the pollution problem connected with tomato processing but
also to enrich ROO with bioactive molecules, namely lycopene, hence
increasing its shelf-life, nutritional value and health benefits.
2. Material and methods
2.1. Samples
Tomato by-products, peels and seeds, were recovered at the
outlet of a tomato industry located in Siliana, Tunisia, and
dried under sunlight until a dry weight of 95 ± 2%. The sun-
dried tomato by-products were sieved manually to separate
tomato seeds from tomato peels resulting in 35% and 65% on
dry weight basis, respectively. Then, tomato peels were finely
ground using house grinder and stored at −20 ◦ C until further
utilization (Kehili et al., 2017a).
The refined olive oil (ROO) was purchased from a local olive
oil refining industry (AGRO-ZITEX, Sfax) for the maceration
extraction experiments.
2.2. Reagents and standards
Lycopene standard was purchased from Sigma Chemical Co.
(St. Louis, MO, USA). All used solvents were of HPLC grade.
2.3. Total lycopene quantification in tomato peels
by-product
The total extractable lycopene content in tomato peels by-
product was determined using the hexane Soxhlet extraction
for 12 h (Roh et al., 2013). The Soxhlet extraction was carried
out in triplicate and the extracts, also named oleoresins, were
dried using a rotary evaporator. The oleoresin mass was quan-
tified gravimetrically, then dissolved in ethyl acetate, filtrated
using 0.22 ␮m hydrophobic PTFE membrane filters and the
total lycopene content (considered as 100% extraction yield)
was determined using high performance liquid chromatogra-
phy (HPLC) as detailed below (Huang et al., 2008; Roh et al.,
2013).
2.4. Lycopene extraction from tomato peels by-product
by maceration in refined olive oil
The experiments consisted of the maceration of tomato
peels within 30 mL of refined olive oil (ROO) using a hot-
plate magnetic stirrer (Lab Tech, LMS-1003, Korea) for 1 h.
A Box–Behnken response surface design was applied to
maximize the lycopene recovery within ROO under variable
temperatures of 40–80 ◦ C, agitation speeds of 200–400 rpm and
tomato peels to ROO ratios of 2.5–5.5% (w/v). The Box–Behnken
design (BBD) was selected specifically because it requires
fewer runs than the central composite design (CCD) in the
cases of three or four variables and because it is useful and
practically feasible in the case of our study (Prakash Maran
et al., 2013; Jalili et al., 2017). The ranges of the maceration
parameters were preliminarily determined on the basis of
single-factor experiments for the extraction of lycopene from
tomato peels within ROO.
Results from this BBD were analyzed using Design Expert
7.0 software (Stat-Ease, Inc., USA). To determine the optimum
extraction time, the same experiment was performed at vari-
able maceration times ranging from 15 to 150 min under the
optimum conditions of the advanced BBD.
2.5. Lycopene quantification using high performance
liquid chromatography (HPLC)
Lycopene-rich ROO samples were dissolved in ethyl acetate
and analyzed using HPLC-DAD instrument (Agilent, Wald-
bronn, Germany) for lycopene quantification. The HPLC
system was equipped with a ZORBAX Eclipse XDB-C18 column
(3.5 ␮m; 4.6 mm × 250 mm). The chromatographic analysis
was carried out using an isocratic elution mode with Ace-
tonitrile/Dichloromethane mixture (75:25; v/v) applied at a
constant flow rate of 1.5 mL/min and measured at a wave-
length of 470 nm. Lycopene was analyzed in the oil samples
 Food and Bioproducts Processing 1 1 7 ( 2 0 1 9 ) 321–328
by comparison with the retention time and peak area of the
lycopene standard analyzed under the same conditions (Kehili
et al., 2016).
2.6. Statistical analysis
Design Expert 7.0 software (Stat-Ease, Inc., USA) was used for
the generation and evaluation of the statistical Box–Behnken
experimental design. All the statistical analyses and the effect
of variables and their interactions were assessed by analysis of
variance (ANOVA). Differences were considered significant at a
confidence level superior to 95% (p < 0.05). The fitting accuracy
of the model was evaluated by analyzing the regression coeffi-
cient (R2 ). The statistical significance of the model parameters
were evaluated by F-test and p-values. The three dimensional
(3D) response surface plots and two dimensional (2D) contour
plots were analyzed to depict the effect of variables interac-
tion on the measured response, being the lycopene extraction
yield.
3. Results and discussion
3.1. Tomato by-products collect and pretreatment
The ratio of the wet mass of tomato industrial by-products was
determined with reference to the mass of the processed fresh
tomatoes within the same working days. The average ratio
over different working days was of 2.22 ± 0.58%. This value is
comparable with that reported in literature. Indeed, accord-
ing to Del Valle et al. (2006), tomato-processing by-products
consisting of peels and seeds, also called tomato pomace, rep-
resented around 4% of the fruit weight. Similarly, the study by
Zuorro et al. (2012) showed that tomato pomace constitutes
about 3–5% of the total weight of processed tomatoes which,
according to the estimates provided by the World Processing
Tomato Council (WPTC, 2013), leads to an annual production
quantity around 1.5 million tons of tomato by-products world-
wide.
At the outlet of the tomato processing plant, tomato
pomace had a water content of 60.57 ± 2.61%. Therefore, these
by-products were sundried to a dry content of 95 ± 2%, then
manually separated into peels and seeds fractions consisting
of 65% and 35% of the dry pomace, respectively (Table 1).
3.2. Total lycopene content in tomato peels by-product
The hexane Soxhlet extraction showed that the total
lycopene content in tomato peels by-product was equal to
1244.50 ± 30.41 mg/kg, on dry basis (Table 1). Interestingly, the
present result is comparable with that reported in our pre-
vious study by Kehili et al. (2017b) where the initial lycopene
content in tomato peels by-product was of 1198 ± 71.86 mg/kg.
In the following, lycopene extraction yields are expressed as
Table 1 – Characterization of tomato industrial
by-products.
Parameter Tomato peels Tomato seeds
Water content (%) 60.57 ± 2.61
Dry matter after sun drying (%) 95.00 ± 2.05
Proportion (% d.w) 65.02 ± 1.68 34.98 ± 1.68
Lycopene (mg/kg d.w) 1244.5 ± 30.41 33.86 ± 1.62
d.w refers to dry weight.
323
percentages of this maximum value. Remarkably, using the
Soxhlet extraction and the same quantification method as for
lycopene from tomato peels, the tomato seeds fraction was
shown to contain an amount of lycopene equal to 33.86 ± 1.62
(Table 1). As carotenoids are extractable within the oil frac-
tion, tomato seeds oil seemed worth investigation as a natural
source of lycopene that is endowed with the ability to protect
from aging, eczema, and UV damage to the skin (Müller et al.,
2013).
3.3. Maceration of tomato peels industrial by-product
in refined olive oil
3.3.1. Box–Behnken design of experiments
As carotenoids are lipophilic pigments, they are highly solu-
ble in fats and organic solvents (Müller et al., 2011). Therefore,
this study considered the refined olive oil (ROO) as an extrac-
tion solvent for the recovery of carotenoids from tomato peels
industrial by-product towards the joint conservation of the oil
and the carotenoids, particularly lycopene. Interestingly, the
study by Chen et al. (2009) revealed that the lycopene isomer-
ization and degradation processes are less pronounced in the
oil-based samples as compared to water-based samples. Effec-
tively, oil and hydrophobic fats tend to “coat” lycopene and
interfere with the ability of oxygen to react with it, hence pre-
venting its possible oxidation and degradation reactions (Chen
et al., 2009).
Practically, a Box–Behnken design was suggested to opti-
mize the lycopene extraction yield using maceration in
30 mL of ROO under variable temperatures (40–80 ◦ C), agita-
tion speeds (200–400 rpm) and tomato peels to oil, S/L, ratios
(2.5–5.5%, w/v) as summarized in Table 2. A total number of 15
batch runs were derived from this Box–Behnken design includ-
ing three center points used to determine the experimental
error.
3.3.2. Model fitting, statistical analysis and parameters
effects
Table 3 presents the analysis of variance (ANOVA) of the
fitted model to the experimental results obtained from the
Box–Behnken design detailed in Table 2. Actually, each param-
eter effect on the response was considered as significant
if the probability of its coefficient in the model was lower
than the usual p-value equal to 0.05. Accordingly, the tem-
perature and the peels to oil (S/L) ratio were deduced to
be highly influencing the lycopene extraction yield within
ROO regarding their respective p-values equal to 0.0017 and
0.0003, whereas the effect of the agitation speed seemed
less significant (Table 3). Besides, all the 2-way interactions
of the three parameters were not significant in the applied
range of values, while the square effects of temperature
and agitation speed seemed to significantly influence the
lycopene extraction yield within ROO, regarding their low p-
values.
Overall, the high correlation coefficient (R2 ), being of 0.97,
revealed that the fitted model adequately represented the
experimental results (Fig. 1). Effectively, the F-value and
p-value of the model were of 21.16 and 0.0018 (p < 0.05),
respectively, which indicates that the model was significant.
Furthermore, the insignificant lack of fit may also strengthen
the reliability of the model as highlighted by Poojary and
Passamonti (2015) (Table 2).
The following equation consists of the fitted prediction
model describing the empirical relationship between the
324 Food and Bioproducts Processing 1 1 7 ( 2 0 1 9 ) 321–328

Table 2 – Box–Behnken design matrix and experimental results of lycopene extraction yields from tomato peels
by-product using maceration in refined olive oil.
Run Temperature (◦ C) Agitation speed (rpm) S/L ratio (%, w/v) Lycopene yield (%)

1 40 200 4 73.03
2 80 200 4 84.34
3 40 400 4 66.15
4 80 400 4 80.92
5 40 300 2,5 73.96
6 80 300 2,5 88.65
7 40 300 5,5 52.41
8 80 300 5,5 66.17
9 60 200 2,5 76.16
10 60 400 2,5 80.19
11 60 200 5,5 60.63
12 60 400 5,5 57.91
13 60 300 4 60.20
14 60 300 4 52.82
15 60 300 4 57.03

Table 3 – Analysis of variance (ANOVA) of the fitted model to the Box–Behnken design results of lycopene extraction yield
from tomato peels by-product as function of the parameters of maceration in refined olive oil.
Source Sum of squares D.F Mean square F-Value p-Value Significance
*
Model 1896,66 9 210,74 21,16 0.0018
A-Temperature (◦ C) 371,69 1 371,69 37,32 0.0017 *

B-Agitation (rpm) 10,10 1 10,10 1,01 0.3601

**
C-Ratio S/L (%, w/v) 837,22 1 837,22 84,06 0.0003
A*B 2,99 1 2,99 0,30 0.6071
A*C 0,22 1 0,22 0,02 0.8886
B*C 11,39 1 11,39 1,14 0.3338
Aˆ2 407,14 1 407,14 40,88 0.0014 *

Bˆ2 294,17 1 294,17 29,54 0.0029 *

Cˆ2 35,79 1 35,79 3,59 0.1165

Residual 49,80 5 9,96
Lack of fit 22,39 3 7,46 0,54 0.6986
Pure error 27,41 2 13,71
Corrected total 1946,46 14

D.F refers to the degree of freedom.

∗
Significant at a confidence level of 95%.
∗∗
Significant at a confidence level of 99.9%.

response, being the lycopene extraction yield, and the tested

variables:

Lycopeneextractionyield(%) = 256, 29 − 2, 91 × A − 0, 53

× B − 14, 05 × C + 0, 0004 × A × B − 0, 0078 × A × C

− 0, 0113 × B × C + 0, 0263 × Aˆ2 + 0, 0009 × Bˆ2 + 1, 38 ∗ Cˆ2 (1)

with A: Temperature (◦ C), B: Agitation speed (rpm) and C:

S/L ratio (%, w/v).
In the present study, the effects of the maceration process
factors (temperature, S/L ratio, and agitation speed) on the
lycopene extraction yield were investigated using three factors
Fig. 1 – Prediction versus actual plot of the lycopene
at three levels BBD. The 3D response surface plots and the 2D
extraction yields from tomato peels by-product using
contour plots showed the relative effects of each two variables
maceration in refined olive oil according to the quadratic
on the lycopene extraction yield when the remaining variable
model (1).
was kept as constant (Fig. 2). Noticeably, the lycopene extrac-
tion yield increased with the increase of the temperature and
the agitation speed levels while it decreased with the increase 2014). For instance, a high correlation between the lycopene
of the ratio S/L (Fig. 2). Actually, the temperature was high- recovery yield from tomato juice and the extraction tempera-
lighted as an important extraction parameter independently ture using SC-CO2 extraction method was deduced by Egydio
of the extraction methods (Ganje et al., 2018; Rostami et al., et al. (2010). Indeed, in the latter study, the lycopene extraction
 Food and Bioproducts Processing 1 1 7 ( 2 0 1 9 ) 321–328 325

Fig. 2 – 3D response surface plots (A, B, C) and 2D contour plots (A’, B’, C’) of the lycopene extraction yields as function of the
different interactions of the extraction parameters, namely the temperature (◦ C), agitation speed (rpm) and the peels to oil
ratio (S/L, % (w/v)), using maceration in refined olive oil.

yields ranged between 7.7% to 76.9% depending basically on
the temperature level where the highest lycopene yield, 76.9%,
was reached at 80 ◦ C and 350 bar. Effectively, an increase in
the temperature level is known to increase the mass transfer
and the solvent diffusion through the ground vegetable matrix
leading to a higher lycopene recovery yield (Machmudah et al.,
2012; Goula, 2013). A similar effect was noticed for the macer-
ation speed as compared to the maceration temperature since
the raise of the agitation speed seemed to enhance the pen-
etration of the solvent, ROO, into the tomato peels’ tissues
and accelerate the release of the mediated solutes, particu-
larly lycopene, into the solvent by disrupting the cell walls
(Goula, 2013).
Remarkably, the lycopene extraction yields varied between
52.41% and 88.65% as function of the different experimental
conditions, referring to the extraction of 652.24 and 1103.25 mg
lycopene/kg of dry tomato peels, respectively (Table 2). This
broad range of results proved that the investigated values of
the experimental parameters had important effects on the
lycopene extraction yields (Poojary and Passamonti, 2015).
326 Food and Bioproducts Processing 1 1 7 ( 2 0 1 9 ) 321–328
3.3.3. Determination and validation of optimum
conditions
Regarding the experimental results of the Box–Behnken
design, the highest lycopene extraction yield was equal
to 88.65% of the initial lycopene content, being of
1244.50 ± 30.41 mg/kg. This maximum extraction yield
resulted from experiment 6 in Table 2 which was performed
at 80 ◦ C, 300 rpm and 2.5% of tomato peels to oil ratio (w/v).
Based on the prediction model cited above, the maximum
predictable lycopene extraction yield, being equal to 98.33%,
can be reached using the maceration of tomato peels into
ROO at a ratio of 2.5% (w/v) under 400 rpm and 80 ◦ C for 1 h.
To confirm the validity of the empirical model and the
optimum conditions, few experiments were carried out to
compare the experimental and the predicted results of
lycopene extraction yields using the model equation (1). For
instance, under the same optimum operating conditions, the
predicted lycopene extraction yield, 98.33%, agreed closely
with the experimental yield that was equal to 94.09 ± 2.77%.
Hence, the results from the confirmation experiments proved
the suitability of the developed quadratic model and its valid-
ity within the investigated range of the process parameters’
levels.
Importantly, to the best of our knowledge, the maximum
lycopene extraction yield reported herein, 98.33%, referring to
the extraction of 1223.72 ± 29.90 mg/kg of dry tomato peels,
is higher than in most of the previous studies for lycopene
extraction from tomato products (López-Cervantes et al., 2013;
Machmudah et al., 2012; Kehili et al., 2017b). For instance, our
previous study by Kehili et al. (2017b) reported a maximum
lycopene yield equal to 60.85%, referring to the extraction
of 728.98 ± 31.17 mg/kg of dry tomato peels, obtained by SC-
CO2 extraction under 400 bar, 80 ◦ C and 4 g CO2 /min relating
to a solvent to peels ratio of 0.4 g CO2 /g min, for an extrac-
tion time of 105 min. The same study considered lycopene
extraction from tomato peels industrial by-product using con-
ventional maceration extraction using hexane, ethyl acetate
and ethanol (Kehili et al., 2017b). Among the three applied sol-
vents, hexane was deduced as the most convenient solvent for
lycopene extraction, 50.83%, while ethanol and ethyl acetate
resulted in lower lycopene recovery yields, being of 23.75% and
26.74%, respectively. Accordingly, the herein suggested extrac-
tion method using maceration in ROO resulted in significantly
higher lycopene recovery yield, 94.09%, as compared to that
obtained by SC-CO2 extraction, 60.85%, and hexane macera-
tion, 50.83%, starting from the same tomato peels industrial
by-product.
3.3.4. Effect of the maceration time on the lycopene
extraction yield under optimum maceration conditions
Fig. 3 highlights the effect of the maceration time on the
lycopene extraction yield from tomato peels within ROO.
These experiments were run under the optimum maceration
conditions as deduced from the previous BBD, being of 80 ◦ C,
400 rpm and a ratio of biomass to oil of 2.5% (w/v). Impor-
tantly, it is interesting to determine the appropriate extraction
conditions not only for better process optimization and cost-
effectiveness but also for better oil stability and lycopene
bioactivity preservation (Bakhshabadi et al., 2017; Farzaneh
and Carvalho, 2015; Ghodsvali et al., 2016). Obviously, regard-
ing Fig. 3, an extraction time of 45 min was sufficient to extract
99.3 ± 0.62% of the initial lycopene content in tomato peels.
Remarkably, the lycopene extraction yield tended to decrease
beyond an extraction time of 45 min. The lower yield of
Fig. 3 – Effect of different maceration times on the lycopene
extraction yields by maceration of tomato peels in refined
olive oil under optimum extraction conditions, a
temperature of 80 ◦ C, an agitation speed of 400 rpm and a
ratio S/L of 2.5% (w/v).
lycopene obtained at higher extraction time might be caused
by an extensive isomerization and degradation of lycopene
under the high operating temperature, 80 ◦ C, as carotenoids
are sensitive to the high extraction temperatures (Chen et al.,
2009; Ganje et al., 2018). Similar results have been also high-
lighted by other studies as summarized in the review by Strati
and Oreopoulou (2014).
Presently, the highest lycopene extraction yield obtained
using maceration of tomato peels within ROO, 99.3 ± 0.62%,
was higher than in most of the previous studies using super-
critical CO2 and solvent extraction of lycopene from tomatoes
products and wastes. For instance, the present result is higher
than the optimal lycopene extraction yield, being of 75.75%,
obtained by Poojary and Passamonti (2015) using acetone/n-
hexane (1:3, v/v) solvent extraction at a temperature of 30 ◦ C
from tomato processing wastes, after kinetic and mass trans-
fer modeling of variable extraction conditions, namely the
temperature and solvent composition.
The present optimum lycopene extraction yield, 99.3%, cor-
responded to a lycopene concentration equal to 34.11 mg/kg
of ROO, being the extraction solvent. Interestingly, lycopene
which is well-known for its strong antioxidant power is
assumed to strongly protect ROO against the primary oxi-
dation reactions as previously highlighted in our study by
Kehili et al. (2017a). Indeed, in the latter study, the efficacy of
lycopene rich-tomato peels oleoresin (TPO) was investigated
in stabilizing refined olive (ROO) and sunflower (RSO) oils for
five months, under accelerated shelf-life, and compared to the
synthetic antioxidant, butylated hydroxytoluene (BHT). The
stability of oils was evaluated by measuring the peroxide value
(PV), acidity and conjugated dienes and trienes (K232 and K270 )
(Kehili et al., 2017a).
In comparison with the present results, a quite similar
lycopene concentration equal to 40 mg/kg in ROO, reported in
our previous study by Kehili et al. (2017a), led to comparable oil
stability indexes as ROO-BHT samples, referring to ROO sam-
ples supplemented with BHT at its legal concentration limit
being of 200 mg/kg, during 7 weeks of the accelerated shelf
life test. Unexpectedly, from week 11 up to week 19, corre-
sponding to the end of the experimental period, the oxidative
stability indexes were lower in ROO samples supplemented
with 40 and 20 mg lycopene/kg oil as compared to the control
sample (ROO-Ctrl), without any antioxidant supplementation,
and ROO-BHT samples. Meanwhile, lower lycopene concen-
tration consisting of 5 and 10 mg/kg ROO resulted in higher
oil stability as compared to ROO-Ctrl and ROO-BHT samples,
during all the experimental period. This effect was assumed
 Food and Bioproducts Processing 1 1 7 ( 2 0 1 9 ) 321–328
as the prooxidation effect of lycopene at high concentration
and after a long storage period. Effectively, previous studies
exhibited that the prooxidation phenomenon can be initiated
by an autooxidation of the natural antioxidants or by their
reaction with free radicals and polyunsaturated fatty acids
(Eder et al., 2002). For instance, lycopene, ␤-carotene and vita-
min E, among other antioxidants, were reported to display
prooxidative effects when applied at high concentrations or
undertaken to harsh operating conditions such as thermally
oxidized bulk oil systems or cells (Kehili et al., 2017a; Park et al.,
2013). With regard to this issue, an intriguing solution may
consist of the edible consumption of the lycopene-enriched
ROO within a short period of time to avoid the prooxidation
phenomenon occurring after long storage periods. Still, we
cannot deny that a compromise should be made between the
extraction time, hence the energy consumption for agitation
and heating, and the lycopene extraction yield and stability.
In addition, in the same study, a statistical analysis was per-
formed to discern the correlation models relating the refined
oils stability indexes with their lycopene contents (Kehili et al.,
2017a). Consequently, we deduced that the oxidative stabil-
ity indexes, particularly the primary oxidation indexes, PV
and K232 , were significantly correlated with the lycopene con-
tent in ROO and RSO samples (Kehili et al., 2017a). Indeed,
it was noticeable that the higher the lycopene conservation
rate is, the lower is the production of primary oxidation com-
pounds and the higher is the refined oil stability. The relation
between the oil stability and the lycopene preservation was
also reported in the studies by Anguelova and Warthesen
(2000) and Boon et al. (2008).
To sum up, the present study provides new insights on the
lycopene extraction from tomato peels industrial by-product
by maceration in ROO. However, further studies are required
towards the development of an industrial extraction process,
including more in depth studies to optimize the number of
sequential maceration steps for the enrichment of the extract-
ing oil in lycopene and enhance the efficacy of the large-scale
extraction system. Besides, as the extraction conditions have
very substantial impacts not only on the lycopene extraction
yield, but also on the bioactivity of the achieved lycopene-
enriched oil samples (Farzaneh et al., 2018; Bakhshabadi et al.,
2017; Farzaneh and Carvalho, 2017; Jalili et al., 2017), next study
should consider the effects of the maceration conditions on
the antioxidant activity of the resulting lycopene-rich ROO.
Therewith, much in depth work is required for the identifica-
tion of all the active components in TPO and the assessment of
the mechanisms of their individual contribution to the antiox-
idative activity in refined oils under long storage periods.
4. Conclusions
In the present study, the refined olive oil (ROO) was used as an
extraction solvent for the recovery of carotenoids, particularly
lycopene, from tomato peels industrial by-product. Results
from the Box–Behnken design allowed to develop a significant
quadratic model for the prediction of lycopene extraction yield
as function of tomato peels maceration parameters. Under
optimum maceration conditions consisting of a temperature
of 80 ◦ C, a biomass to oil ratio of 2.5% (w/v), a magnetic stir-
ring of 400 rpm for 45 min, almost the total lycopene content
was recovered from tomato peels by-product leading to a final
concentration of 35 mg/kg in ROO. Hence, this study revealed
that tomato peels is a potential source of lycopene and its
327
extraction using maceration in ROO seems an efficient and
economic method not only for the recovery of the total amount
of lycopene, but also for the mutual stabilization of lycopene
and ROO against potential peroxidation during long storage
periods.
Conflicts of interest
The authors declare no conflict of interest.
Acknowledgments
The study was performed as part of a MOBIDOC thesis funded
by the European Union within the PASRI program adminis-
tered by the “Agence Nationale de Promotion de la Recherche
(ANPR)”. The work was also funded by the Ministry of Higher
Education and Scientific Research, Tunisia.
References
Amiri-Rigi, A., Abbasi, S., 2016. Microemulsion-based lycopene
extraction: effect of surfactants, co-surfactants and
pretreatments. Food Chem. 197, 1002–1007.
Anguelova, T., Warthesen, J., 2000. Degradation of lycopene,
␣-carotene, and ␤-carotene during lipid peroxidation. J. Food
Sci. 65 (1), 71–75.
Baccouri, O., Guerfel, M., Baccouri, B., Cerretani, L., Bendini, A.,
Lercker, G., Zarrouk, M., Daoud Ben Miled, D., 2008. Chemical
composition and oxidative stability of Tunisian monovarietal
virgin olive oils with regard to fruit ripening. Food Chem. 109,
743–754.
Bakhshabadi, H., Mirzaei, H., Ghodsvali, A., Jafari, S.M., Ziaiifar,
A.M., Farzaneh, V., 2017. The effect of microwave
pretreatment on some physico-chemical properties and
bioactivity of Black cumin seeds’ oil. Ind. Crops Prod. 97, 1–9.
Ben-Ali, M., Dhouib, K., Damak, M., Allouche, N., 2014.
Stabilization of sunflower oil during accelerated storage: use
of basil extract as a potential alternative to synthetic
antioxidants. Int. J. Food Prop. 17 (7), 1547–1559.
Boon, C.S., Xu, Z., Yue, X., McClements, D.J., Weiss, J., Decker, E.A.,
2008. Factors affecting lycopene oxidation in oil-in-water
emulsions. J. Agric. Food Chem. 56 (4), 1408–1414.
Chen, J., Shi, J., Xue, S.J., Mac, Y., 2009. Comparison of lycopene
stability in water- and oil-based food model systems under
thermal- and light-irradiation treatments. LWT-Food Science
and Technology 42, 740–747.
Del Valle, M., Camara, M., Torija, M., 2006. Chemical
characterization of tomato pomace. J. Sci. Food Agric. 86 (2),
1232–1236.
Dolatabadi, Z., Elhami Rad, A.H., Farzaneh, V., Akhlaghi Feizabad,
S.H., Estiri, S.H., Bakhshabadi, H., 2016. Modeling of the
lycopene extraction from tomato pulps. Food Chem. 190,
968–973.
Eder, K., Flader, D., Hirche, F., Brandsch, C., 2002. Excess dietary
Vitamin E lowers the activities of antioxidative enzymes in
erythrocytes of rats fed Salmon oil. J. Nutr. 132, 3400–3404.
Egydio, J.A., Moraes, Â.M., Rosa, P.T.V., 2010. Supercritical fluid
extraction of lycopene from tomato juice and characterization
of its antioxidation activity. J. Supercrit. Fluids 54, 159–164.
Farzaneh, V., Gominho, J., Pereira, H., Carvalho, I.S., 2018.
Screening of the antioxidant and enzyme inhibition
potentials of Portuguese Pimpinella anisum L. seeds by GC-MS.
Food Anal. Methods 11 (10), 2645–2656.
Farzaneh, V., Carvalho, I.S., 2017. Modelling of microwave
assisted extraction (MAE) of anthocyanins (TMA). J. Appl. Res.
Med. Aromat. Plants 6, 92–100.
Farzaneh, V., Carvalho, I.S., 2015. A review of the health benefit
potentials of herbal plant infusions and their mechanism of
actions. Ind. Crops Prod. 65, 247–258.
328 Food and Bioproducts Processing 1 1 7 ( 2 0 1 9 ) 321–328
Galanakis, C.M., 2013. Emerging technologies for the production
of nutraceuticals from agricultural by-products: a view point
of opportunities and challenges. Food Bioprod. Process. 91,
575–579.
Ganje, M., Jafari, S.M., Farzaneh, V., Malekjani, N., 2018. Kinetics
modelling of color deterioration during thermal processing of
tomato paste with the use of response surface methodology.
Heat Mass Transf. 54, 3663–3671.
García, A., Ruiz-Méndez, M.V., Romero, C., Brenes, M., 2006. Effect
of refining on the phenolic composition of crude olive oils. J.
Am. Oil Chem. Soc. 83 (2), 159–164.
Ghodsvali, A., Farzaneh, V., Bakhshabadi, H., Zare, Z., Karami, Z.,
Mokhtarian, M., Carvalho, I.S., 2016. Screening of the
aerodynamic and biophysical properties of barley malt. Int.
Agrophys. 30 (4), 457–464.
Goula, A.M., 2013. Ultrasound-assisted extraction of pomegranate
seed oil – kinetic modeling. J. Food Eng. 117, 492–498.
Huang, W., Li, Z., Niu, H., Li, D., Zhang, J., 2008. Optimization of
operating parameters for supercritical carbon dioxide
extraction of lycopene by response surface methodology. J.
Food Eng. 89, 298–302.
Issaoui, M., Flamini, G., Brahmi, F., Dabbou, S., Ben Hassine, K.,
Taamali, A., Chehab, H., Ellouze, M., Zarrouk, M., Hammami,
M., 2010. Effect of the growing area conditions on
differentiation between Chemlali and Chétoui olive oils. Food
Chem. 119 (1), 220–225.
Jalili, F., Jafari, S.M., Emam-Djomeh, Z., Malekjani, N., Farzaneh,
V., 2017. Optimization of ultrasound-assisted extraction of oil
from canola seeds with the use of response surface
methodology. Food Anal. Methods 11, 598–612.
Kehili, M., Choura, S., Zammel, A., Allouche, N., Sayadi, S., 2017a.
Oxidative stability of refined olive and sunflower oils
supplemented with lycopene-rich oleoresin from tomato
peels industrial by-product, during accelerated shelf-life
storage. Food Chem. 246, 295–304.
Kehili, M., Kammlott, M., Choura, S., Zammel, A., Zetzl, C.,
Smirnova, I., Allouche, N., Sayadi, S., 2017b. Supercritical CO2
extraction and antioxidant activity of lycopene and
␤-carotene-enriched oleoresin from tomato (Lycopersicum
esculentum L.) peels by-product of a Tunisian industry. Food
Bioprod. Process. 102 (2), 340–349.
Kehili, M., Schmidt, L.M., Reynolds, W., Zammel, A., Zetzl, C.,
Smirnova, I., Allouche, N., Sayadi, S., 2016. Biorefinery cascade
processing for creating added value on tomato industrial
by-products from Tunisia. Biotechnol. Biofuels 261 (9), 1–12.
Khlebnikov, A.I., Schepetkin, I.A., Domina, N.G., Kirpotina, L.N.,
Quinn, M.T., 2007. Improved quantitative structure–activity
relationship models to predict antioxidant activity of
flavonoids in chemical, enzymatic, and cellular systems.
Bioorg. Med. Chem. 15, 1749–1770.
Lianfu, Z., Zelong, L., 2008. Optimization and comparison of
ultrasound/microwave assisted extraction (UMAE) and
ultrasonic assisted extraction (UAE) of lycopene from
tomatoes. Ultrason. Sonochem. 15, 731–737.
López-Cervantes, J., Sánchez-Machado, D.I., Valenzuela-Sánchez,
K.P., Núñez-Gastélum, J.A., Escárcega-Galaz, A.A.,
Rodríguez-Ramírez, R., 2013. Effect of solvents and methods
of stirring in extraction of lycopene, oleoresin and fatty acids
from over-ripe tomato. Int. J. Food Sci. Nutr. 65 (2),
187–193.
Machmudah, S., Zakaria, Winardi, S., Sasaki, M., Goto, M.,
Kusumoto, N., Hayakawa, K., 2012. Lycopene extraction from
tomato peel by-product containing tomato seed using
supercritical carbon dioxide. J. Food Eng. 108 (2), 290–296.
Moure, A., Cruz, J.M., Franco, D., Domínguez, J.M., Sineiro, J.,
Domínguez, H., Nunez, M.J., Parajo, J.C., 2001. Natural
antioxidants from residual sources. Food Chem. 72 (2),
145–171.
Müller, L., Catalano, A., Simone, R., Cittadini, A., Fröhlich, K.,
Böhm, V., Palozza, P., 2013. Antioxidant capacity of tomato
seed oil in solution and its redox properties in cultured
macrophages. J. Agric. Food Chem. 61 (2), 346–354.
Müller, L., Fröhlich, K., Böhm, V., 2011. Comparative antioxidant
activities of carotenoids measured by ferric reducing
antioxidant power (FRAP), ABTS bleaching assay (␣TEAC),
DPPH assay and peroxyl radical scavenging assay. Food Chem.
129 (1), 139–148.
Papaioannou, E.H., Kyriakides, M.L., Karabelas, A.J., 2015. Natural
origin lycopene and its “green” downstream processing. Crit.
Rev. Food Sci. Nutr. 56 (4), 686–709.
Papaioannou, E.H., Karabelas, A.J., 2012. Lycopene recovery from
tomato peel under mild conditions assisted by enzymatic
pre-treatment and non-ionic surfactants. Acta Biochim. Pol.
59 (1), 71–74.
Park, J., Kim, T.S., Kim, M.J., Lee, J., 2013. Prooxidative and
antioxidative properties of ␤-carotene in chlorophyll and
riboflavin photosensitized oil-in-water emulsions. Food
Chem. 140 (1–2), 255–261.
Poojary, M.M., Passamonti, P., 2015. Extraction of lycopene from
tomato processing waste: kinetics and modelling. Food Chem.
173, 943–950.
Prakash Maran, J., Manikandan, S., Mekala, V., 2013. Modeling and
optimization of betalain extraction from Opuntia ficus-indica
using Box–Behnken design with desirability function. Ind.
Crops Prod. 49, 304–311.
Roh, M.-K., Jeon, M.-H., Moon, J.-N., Moon, W.-S., Park, S.-M., Choi,
J.-S., 2013. A simple method for the isolation of lycopene from
Lycopersicon esculentum. Bot. Sci. 91 (2), 187–192.
Rostami, M., Farzaneh, V., Boujmehrani, A., Mohammadi, M.,
Bakhshabadi, H., 2014. Optimizing the extraction process of
sesame seed’s oil using response surface method on the
industrial scale. Ind. Crops Prod. 58, 160–165.
Shan, B., Xie, J., Zhu, J., Peng, Y., 2011. Ethanol modified
supercritical carbon dioxide extraction of flavonoids from
Momordica charantia L. and its antioxidant activity. Food
Bioprod. Process. 90, 579–587.
Strati, I.F., Gogou, E., Oreopoulou, V., 2015. Enzyme and high
pressure assisted extraction of carotenoids from tomato
waste. Food Bioprod. Process. 94, 668–674.
Strati, I.F., Oreopoulou, V., 2014. Recovery of carotenoids from
tomato processing by-products – a review. Food Res. Int. 65,
311–321.
Torbica, A., Belovic, M., Mastilovic, J., Kevresan, Z., Pestoric, M.,
Skrobot, D., Hadnadev, T.D., 2016. Nutritional, rheological, and
sensory evaluation of tomato ketchup with increased content
of natural fibres made from fresh tomato pomace. Food
Bioprod. Process. 98, 299–309.
WPTC (The World Processing Tomato Council), 2013. World
Production Estimate as of 25 October 2013 (Accessed 13
December 2013) www.wptc.to.
Xu, Y., Pan, S., 2013. Effects of various factors of ultrasonic
treatment on the extraction yield of all-trans-lycopene from
red grapefruit (Citrus paradise Macf.). Ultrason. Sonochem. 20,
1026–1032.
Zuorro, A., Lavecchia, R., Medici, F., Piga, L., 2012.
Enzyme-assisted production of tomato seed oil enriched with
lycopene from tomato pomace. Food Bioprocess Technol. 6
(12), 3499–3509.