J.

of Supercritical Fluids 72 (2012) 168–175

Contents lists available at SciVerse ScienceDirect

The Journal of Supercritical Fluids

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

Extraction of antioxidant compounds from different varieties of Mangifera indica

leaves using green technologies
Ma Teresa Fernández-Ponce ∗ , Lourdes Casas, Casimiro Mantell, Miguel Rodríguez,
Enrique Martínez de la Ossa
Department of Chemical Engineering and Food Technology, Science Faculty, University of Cádiz International Agri-food Campus of Excellence, ceiA3,
P.O. Box 40, Puerto Real 11510, Cádiz, Spain

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

Article history: Supercritical Fluid Extraction (SFE) and Subcritical Water Extraction (SWE) from mango leaves were
Received 15 May 2012 applied in order to obtain extracts with high phenolic content and potent antioxidant activity. The
Received in revised form 25 July 2012 effects of extraction conditions on sub- and supercritical CO2 extraction were analyzed: temperature
Accepted 26 July 2012
(35 and 55 ◦ C), pressure (10 and 40 MPa), percentage of co-solvent (0 and 20%) and type of co-solvent
(methanol/ethanol). The best condition (CO2 + 20% of ethanol at 10 MPa, 55 ◦ C, 20 g/min and 3 h) was com-
Keywords:
pared with SWE (4 MPa, 100 ◦ C, 10 g/min, and 3 h) using seven mango cultivars. SWE was more efficient
Mango leaves
than subcritical CO2 + ethanol. The antioxidant activity was evaluated by DPPH assay, and the quantifi-
Supercritical Fluid Extraction
Subcritical Water Extraction
cation of the main polyphenols of mango leaves by HPLC analysis. SWE showed global yields up to 35%
Co-solvents for Kent variety, and extracts with antioxidant activities superior to (+)-␣-tocopherol related with their
Mangiferin high content on the polyphenols mangiferin and quercetin.
Quercetin © 2012 Elsevier B.V. All rights reserved.
Antioxidant activity

1. Introduction compounds [4–8]. Both techniques have been widely explored in

recent years in order to recover bioactive compounds from diverse
The challenges of this century based on a sustainable and plants and agri-industrial by-products [4–22].
more friendly environment development have turned the vision Agricultural by-products of mango, particularly leaves and bark,
of chemical production toward a new industry concept of biomass present a high content on potent phenolic compounds, mainly
refining in order to decrease rapid consumption of non-renewable mangiferin and quercetin, whose pharmaceuticals and nutraceu-
resources (petroleum, natural gas, coal, and minerals). In the begin- tics properties have been demonstrated in several studies [23–30].
ning, a typical biorefinery convert essentially natural renewable Mango is one of the most important tropical fruit worldwide
matter into bio-energy products. However, in the next generation with a global production superior to 38 million tones and an
biorefinery, the feedstock will be fractionated further into valuable area harvested superior to 5 million hectares in 2010 [31]. Annu-
components by extraction, fermentation and controlled pyrolysis, ally pruning activity generates considerable quantities of residues
as well as by more traditional methods. which are usually burned or used for soil amelioration. Thus, con-
One of the first stages in the new biorefinery is the extraction version of pruning mango residues into valuable chemical products
of secondary metabolites from low value biomass considering that by efficient and low impact extraction techniques results clearly
they are of greater value in cosmetic, nutraceutical and pharmaceu- attractive within the concept of biorefinery.
tical industries. The use of harmless extraction methods is essential The extraction from mango by-products using SC-CO2 or sub-
to comply with and environmental compatible and sustainable critical water has not been widely studied. Traditional solvent
chemical production [1–3]. extraction techniques are still usually used to recover bioactive
Supercritical Fluid Extraction (SFE) and Subcritical Water compounds from mango [23,25–30] despite the drawbacks present
Extraction (SWE) are interesting alternatives so present sev- in these techniques [8,11–13,16,18].
eral advantages including the use of green solvents, faster and Mango leaves extracts with antioxidant activity have been
more selective processes, and the low degradation of chemical obtained by SC-CO2 extraction [32], but pure CO2 , a nonpolar sol-
vent, provide a low efficiency to extract highly or slightly polar
compounds. Thus, the addition of CO2 modifiers such as alcohol co-
∗ Corresponding author. Tel.: +34 956 016 579; fax: +34 956 016 411. solvents should increase the extraction of polar polyphenols and
E-mail address: [email protected] (M.T. Fernández-Ponce). also improve the antioxidant activity of extracts, as described by

0896-8446/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.supflu.2012.07.016
 M.T. Fernández-Ponce et al. / J. of Supercritical Fluids 72 (2012) 168–175 169

global yield (X0 ) for all extraction method was calculated consider-
ing the ratio between mass of extract and mass of dry raw material.
A preliminary study was conducted in order to improve the yield
and antioxidant activity of the extracts obtained using CO2 and CO2
plus co-solvents. The effects of different variables on the extrac-
tion process were analyzed by considering the following operating
conditions: pressures of 10 and 40 MPa, temperatures of 35 and
55 ◦ C, co-solvent percentages of 0 and 20% and type of co-solvent,
methanol and ethanol. All tests were carried out with a CO2 flow
rate of 20 g/min and an extraction time of 3 h.
Results were compared with SWE. This technique is less depend-
ent on pressure and highly dependent on temperature [19,20].
However, it is important to consider that the temperatures above
100 ◦ C could generate unwanted oxidative processes [22], thus SWE
tests were carried out at 100 ◦ C, 4 MPa, a flow rate of 10 g/min and
Fig. 1. Schematic diagram of the high pressure equipment. 3 h.
For this preliminary study mango leaves of the variety Osteen
other authors using different natural matters [7,8,10–15,18]. On were used as raw material so it is the variety widely cultivated in
the other hand, although SWE is efficient to extract polar com- the region of Málaga, Spain.
pounds, this technique has not been evaluated before using this
raw material. 2.3. Antioxidant activity assay with DPPH
Therefore, in this work Sub- and Supercritical Fluid Extrac-
tion from mango leaves using pure CO2 and CO2 plus co-solvents Antioxidant activity of extracts and standard compounds ((+)-
at different conditions was studied and compared with SWE. ␣-tocopherol, mangiferin and quercetin 3-␤-d-glucoside) was
Extracts were evaluated considering the global extraction yield, the determined by 2,2-diphenyl-1-picrylhydrazyl radical (DPPH) assay.
antioxidant activity and the phenolic composition. In addition, the The method employed was designed having in account the meth-
performance of seven varieties of mango leaves was analyzed using ods described by Brand-Williams and Scherer and Godoy [33,34].
CO2 plus co-solvents and SWE. About 0.1 mL aliquots of methanolic solutions of the samples or
standards at different concentrations were each added to 3.9 mL
2. Materials and methods of a 6 × 10−5 mol/L DPPH methanolic solution. The absorbance of
DPPH was monitored spectrophotometrically at 515 nm at 0 min
2.1. Materials and every 2 min until the reaction reached the steady state. The
DPPH concentration (CDPPH ) in the reaction medium was calculated
The seven varieties of Mangifera indica L. leaves studied (Kens- from a calibration curve determined by linear regression with Eq.
ington, Kent, Keitt, Tommy Atkins, Osteen, Ataulfo and Langra) were (1):
provided by “Estación Experimental La Mayora”, Superior Centre of
Abs = 12, 709 · CDPPH + 0.002 (1)
Scientific Research (CSIC), Málaga, Spain. The leaves were collected
in June 2010 and February 2011. All leaves were dried at room tem- The percentage of DPPH remaining was calculated as described
perature until constant weight and kept frozen in the absence of in Eq. (2)
light.
CDPPHt
Carbon dioxide (99.995%) was provided by Abello-Linde %DPPH remaining = × 100 (2)
CDPPH0
S.A. (Barcelona, Spain). 2,2-Diphenyl-1-picrylhydrazyl, free
radical (DPPH), mangiferin (1,3,6,7-tetrahydroxyxanthone C2- The EC50 (efficient concentration providing 50% inhibition) was
␤-d-glucoside), quercetin 3-␤-d-glucoside, purity ≥ 90% HPLC calculated graphically using a non-linear fitting curve by plotting
grade (3,3 ,4 ,5,7-pentahydroxyflavone 3-␤-d-glucoside), and the sample concentration vs. the % DPPH remaining on steady state.
(+)-␣-tocopherol were provided by Sigma–Aldrich (Steinheim, The antioxidant activity was expressed as the Antioxidant Activity
Germany). The organic solvents ethanol, methanol and acetic acid, Index (AAI) which was calculated considering the final concentra-
all HPLC gradient grade, were provided by Panreac (Barcelona, tion of DPPH and the EC50 of the tested compound in the reaction
Spain). The water used in all experiments was double-distilled as follows (Eq. (3)):
milliQ grade.
final concentration of DPPH (␮g/mL)
AAI = (3)
EC50 (␮g/mL)
2.2. Extraction procedure with solvents at high pressures
The final concentration of DPPH was calculated respect to the
Extraction tests were carried out in a high pressure apparatus concentration of DPPH in the reaction medium. Plant extracts
supplied by Thar Technology (Pittsburgh, PA, USA, model SF100). A showed poor antioxidant activity when AAI < 0.5, moderate antiox-
schematic diagram of the equipment used in this work is shown in idant activity when AAI is between 0.5 and 1.0, strong antioxidant
Fig. 1. This set-up included an extraction vessel (capacity of 100 mL) activity when AAI is between 1.0 and 2.0, and very strong when
with a thermostatic jacket to control the extraction temperature, AAI > 2.0 [34]. The assays were carried out in triplicate. Results
two pumps with a maximum flow rate of 50 g/min (one for car- were compared with standards of (+)-␣-tocopherol, mangiferin and
bon dioxide and the other for co-solvent), a back pressure valve quercetin 3-␤-d-glucoside.
regulator to control the system pressure, and a cyclonic separator
to allow periodic discharge of the extracted material during the 2.4. Identification and quantification of phenolic compounds by
extraction process. For all tests the extraction vessel was loaded HPLC
with approximately 15 g of sample. Extracts were recovered in a
cyclonic separator and then collected in glass bottles and stored Separation of phenolic compounds was performed using an Agi-
in the extraction solvent in darkness at −20 ◦ C prior to assay. The lent HPLC series 1100 system (Agilent, Germany) equipped with a
170 M.T. Fernández-Ponce et al. / J. of Supercritical Fluids 72 (2012) 168–175

quaternary pump, an autosampler, a 250 mm × 4.6 mm i.d., 5 ␮m,

C18 reversed-phase column (Thermo Electron Corporation) and a
UV/vis detector, connected to a HP ChemStation® software.
The method used is a modification of the method described by
Barreto et al. [23]. The solvents that constituted the mobile phase
were A (acetic acid–water, 2:98, v/v) and B (methanol). The elution
conditions applied were: 0–2 min, 5% B isocratic; 2–7 min, linear
gradient 5–25% B; 7–11 min, 25% B isocratic; 11–19 min, linear
gradient 25–32% B; 19–27 min, 32% B isocratic; 27–28 min, linear
gradient 32–40% B; 28–38 min, 40% B isocratic and finally, washing
and reconditioning steps of the column were included (38–50 min,
linear gradient 40–100% B; 50–60 min, 100% B isocratic; 60–70 min,
linear gradient 100–5% B; and 5 min, 5% B isocratic). The flow-rate
was 0.9 ml/min and the injection volume was 50 ␮L. The system
was operated at room temperature.
Phenolic compounds were detected at 340 nm by its retention
times and quantified using a calibration curve of the corre-
sponding standard compounds. The calibration curve of the main
polyphenols of mango leaves, mangiferin (Eq. (4)) and quercetin
3-␤-d-glucoside (Eq. (5)), was as follows:

A = 54, 252 · C − 100.12 (4)

A = 87, 077 · C − 130.11 (5)

where A is the area expressed in mAu and C is the concentration

expressed in ␮g/ml. The correlation coefficient (R2 ) was 0.9999 for
both calibration curves. The experiments on each extraction were
carried out in triplicate in order to evaluate the variability of the
measurements. HPLC chromatograms of the extracts obtained for
two varieties studied are shown in Fig. 5. Fig. 2. Global yield (A) and antioxidant activity (B) of mango leaves extracts of the
variety Osteen obtained using SC-CO2 , CO2 + 20% methanol and CO2 + 20% ethanol.

2.5. Experimental design

preliminary study with Osteen variety: CO2 + 20% (w/w) of ethanol
A multilevel factorial design was carried out in order to
at 10 MPa, 55 ◦ C, 20 g/min and 3 h, and also using subcritical water
determine the effect of temperature and pressure (experimental
at 100 ◦ C, 4 MPa, a flow rate of 10 g/min, and an extraction time of
variables) on the yield of the process when CO2 + 20% of ethanol
3 h. Cultivars were evaluated accordingly to the global extraction
is used as solvent system, accordingly ethanol was selected as the
yield, antioxidant activity and phenolic profile.
best CO2 modifier. The ranges for the factorial design were 35–55 ◦ C
for temperature, and 10–40 MPa for pressure. Then, the extrac-
tion process was analyzed through a factorial experimental design 3. Results and discussion
using 22 + 2 central points where temperature and pressure were
adjusted. On the basis of this design a total of 6 experiments were 3.1. Supercritical Fluid Extraction with pure CO2
carried out in a single block. The response variable was the extrac-
tion yield expressed as g/100 g of dry matter. The experimental One of the advantages of SFE is CO2 solvating power can be
data were analyzed by Statgraphics Plus 5.1® (1994–2001, Statisti- manipulated by changing pressure (P) and/or temperature (T);
cal Graphics Corp.). Empirical correlations were developed in order therefore, knowing the influence of extraction conditions on the
to predict the influence of extraction conditions on the extraction process is necessary to obtain high extraction yields. SC-CO2 extrac-
yield of the process studied. tion yields obtained at different conditions of P and/or T from mango
leaves of the variety Osteen are shown in Fig. 2A. Global yields
2.6. Evaluation of different mango leaves cultivars obtained with pure CO2 were unsatisfactory and even changes in
the extraction conditions led to negligible improvements. The high-
Mango is an Indian native fruit that occupies the third position in est yield was 1.22 ± 0.13% obtained at 40 MPa and 55 ◦ C.
worldwide production and importation of tropical fruits. The cul- The effect of pressure on SC-CO2 extraction from mango leaves
tivars used in the present work were selected according to their was positive for both temperatures studied. This behavior can be
importance worldwide. The Florida cultivars Tommy Atkins, Kent explained by the increase in density with pressure resulting in
and Keitt are the mainly current commercial varieties produced and greater solvating power of CO2 [16,17]. Moreover, high pressures
exported by most countries, including leading exporters as Mexico, cause disruptions in plant cells and allow compounds to be more
Pakistan, Brazil, Peru and Ecuador. Ataulfo is other important cul- available, thereby favoring the extraction yield [37].
tivar of Mexico, which is too appreciated in North America. Osteen About the effect of temperature, it is more complex because it
is the essential cultivar of Spain production. Langra is an important depends on two factors. The density of CO2 decreases with tem-
commercial mango variety of north India with good quality fruits. perature, reducing the solvating power, while the solute vapor
Finally, Kensington is the dominant variety grown in Australia well pressure increases favoring the solubility on SC-CO2 . Thus, depend-
received in overseas markets [35,36]. ing on the operating conditions, one factor is dominant over the
The performance of the M. indica L. varieties above mentioned other [16,17]. In this way, at low pressures (10 MPa) the effect
(Kensington, Kent, Keitt, Tommy Atkins, Osteen, Ataulfo and Lan- on the density was predominant, so high extraction yields were
gra) was evaluated using the best conditions obtained in the obtained on reducing temperature. But, at high pressures (40 MPa)
 M.T. Fernández-Ponce et al. / J. of Supercritical Fluids 72 (2012) 168–175 171

Table 1
Extraction yields of phenolic compounds obtained from mango leaves of the variety Osteen, expressed as mg/100 g dry leaves.

Extraction method Conditions Yield of phenolic compoundsa (mg/100 g dry leaves)

Mangiferin Quercetin
Rt = 20.9 min 3-␤-d-glucoside
Rt = 34.4 min

SC-CO2 100–400 bar 35–55 ◦ C n.d.b n.d.

◦
CO2 + 20% methanol 100 bar 35 C 7.8 ± 0.1 6.1 ± 0.1
55 ◦ C 93.9 ± 0.1 88.7 ± 1.0
400 bar 35 ◦ C 7.8 ± 0.6 6.4 ± 0.1
55 ◦ C 12.9 ± 0.3 15.3 ± 0.1

CO2 + 20% ethanol 100 bar 35 ◦ C 191.8 ± 0.3 140.9 ± 1.2

55 ◦ C 184.5 ± 0.4 231.4 ± 3.3
400 bar 35 ◦ C 62.6 ± 0.2 55.4 ± 0.1
55 ◦ C 40.6 ± 0.0 49.7 ± 0.4

Subcritical water 40 bar 100 ◦ C 1365.9 ± 1.2 409.5 ± 6.7

a
Yield of phenolic compounds expressed as mg/100 g dry matter was represented as the mean value ± standard deviations.
b
n.d.: compound not detected.

the effect of vapor pressure was dominant and the high volatility The extractions with co-solvents were carried out using 20%
of the solutes resulted in an overall increase in the extraction yield. of methanol/ethanol. At the conditions of pressure and tempera-
In spite of the increase on the global yield at high pressure ture tested, the mixtures of CO2 + co-solvent are below their critical
and high temperature, SC-CO2 was not efficient enough to obtain point because high concentrations of CO2 modifiers increase the
high yields from mango leaves. This can be attributing to the high critical point of the mixture [7,39].
content on polar phenolic compounds of mango leaves [23,24], These systems called enhanced fluidity liquids have resulted
poorly soluble in SC-CO2 , a nonpolar solvent [10–14,38]. The main more efficient to extract polar compounds than mixtures with
polyphenols of mango leaves are mangiferin and quercetin 3-␤-d- lower concentrations of co-solvents. For example, other authors
glucoside known by their potent antioxidant properties [24,26–30]. reported that 20% of co-solvents duplicate the extraction yields
Table 1 shows the extraction yields of both compounds obtained of anthocyanins from red grape pomace instead using only 5% of
by HPLC for the different extraction techniques explored, and Fig. 2B co-solvents [7]. Adil et al. [39] showed that the optimum ethanol
shows the antioxidant activity of extracts obtained with SC-CO2 . concentrations for CO2 extraction of polyphenols from apple and
Mangiferin and quercetin 3-␤-d-glucoside were not detected by peach pomace were found to be 20%.
HPLC on SC-CO2 extracts. Both polyphenols are slightly polar com- The reason for this is based on the fact that the solvating power
pounds hence SC-CO2 is not efficient to extract them [38]. As a of CO2 and the solubility of polar compounds in CO2 increase with
result, the antioxidant activity of SC-CO2 extracts was poor too. Any increasing the amount of co-solvent concentration from 5 to 30%,
antioxidant activity was observed for the extracts obtained with SC- due to augmented phenol–alcohol interactions that facilitate the
CO2 at low pressures (10 MPa), and raising the pressure to 40 MPa extraction of the solute [7,38,39].
did not lead to an increase in this activity. Consequently, pure CO2 The global extraction yields and antioxidant activity of extracts
resulted inappropriate to obtain extracts from mango leaves. obtained with both solvent systems are shown in Fig. 2. Comparing
the yield obtained using SC-CO2 and the mixtures of CO2 + 20% of
3.2. Subcritical Fluid Extraction with CO2 plus co-solvents co-solvents (Fig. 2A), it is clear that the addition of methanol or
ethanol allowed higher extraction yields than those obtained with
Given the low capacity of SC-CO2 to extract antioxidant com- pure CO2 at all conditions tested.
pounds from mango leaves, the addition of co-solvents was The influence of P and/or T on the extraction process using sub-
necessary to improve the quality of extracts. The solubility of critical mixtures of CO2 plus co-solvents resulted different than
polar substances, such as polyphenols, in SC-CO2 is very low that observed with pure CO2 . About temperature, the global yields
[10–14,38–40], however the use of organic co-solvents increases increase with temperature when extractions were carried out at
the solvent power of CO2 and the extraction yield [7,18,38–40]. 10 MPa. This positive effect of temperature is related with the
Among all the modifiers including methanol, ethanol, acetoni- increase in the diffusivity and decrease in the viscosity of the
trile, acetone, water, ethyl ether and dichloromethane, methanol solvent improving the mass transfer properties along with the
and ethanol are most frequently used for SFE of polyphenols. Alco- intensification of solute volatility favoring the extraction. However,
hol co-solvents induce dipole/dipole interactions and hydrogen at 40 MPa the effect of temperature on the extraction process from
bonding with polar functional groups, and also they can break polar mango leaves was not relevant using co-solvents.
interactions solute–matrix increasing the solubility of polar solutes In relation to the pressure, at low temperatures (35 ◦ C), pres-
[18,40]. sure had a positive effect on the extraction yields using both
Methanol is commonly used as co-solvent because it is misci- co-solvents. By contrast, at high temperatures (55 ◦ C) a marked
ble up to 20% with CO2 and some publications have shown that increase in the extraction yield was observed when pressure was
this is more efficient than ethanol to remove polyphenols [40,41], decreased from 40 to 10 MPa, showing a negative effect. Low
but the temperature necessary to reach the supercritical state is pressures enhanced the extraction of polyphenols resulting in
higher and could be not suitable for natural products [18]. About extracts with higher content on mangiferin and quercetin and thus
ethanol, it significantly enhanced the extraction of flavonoids [18] better antioxidant capacity. The higher extraction yields for both
and may be a better choice considering it as non-toxic and can be co-solvents (methanol and ethanol) were obtained at 10 MPa and
used in nutraceutical or cosmetic applications [39]. With this in 55 ◦ C (6.53 ± 0.83 and 6.37 ± 0.13%, respectively).
mind, ethanol and methanol were used as CO2 modifiers in this The negative effect of pressure on the extraction process
work. using CO2 + 20% of co-solvent was also observed for the phenolic
172 M.T. Fernández-Ponce et al. / J. of Supercritical Fluids 72 (2012) 168–175

extraction yield. The highest yields of phenolic compounds were

obtained at 10 MPa and 55 ◦ C (184.5 ± 0.4 and 231.4 ± 3.3 mg/100 g
dry leaves for mangiferin and quercetin, respectively) (see Table 1).
This negative effect of pressure has been also described for other
raw materials. Laroze et al. [15] reported that highest phenolic con-
tent was obtained at low pressure, 7.5–15 MPa from cranberry,
blueberry and raspberry. Mantell et al. [7] described an increase
of anthocyanins extraction from grape marc using CO2 + 20%
methanol/water at low pressures. Vatai et al. [14] also observed
that at 15 MPa higher amounts of total phenols were extracted
Fig. 3. Pareto diagram for the global extraction yield obtained with CO2 + 20% of
from grape marc using CO2 plus co-solvents. Finally, Adil et al. [39] ethanol using mango leaves of the variety Osteen.
reported a negative effect of pressure below 50 MPa on total phe-
nolic content of apple and peach pomace using enhanced fluidities.
The Pareto diagram for the global yield as the response variable
A scientific explanation for this behavior is not clear and it is
is shown in Fig. 3. The estimated effects and interactions between
not described in the literature. The above mentioned cases show
the range of variables studied and the analysis of variance are
that lower pressures improve the extraction yield using co-solvents
also shown in Table 2. The degree of significance of each factor
[7,14,15,39]. But other authors reported that high pressures effi-
is represented by its p-value; factors with a p-value < 0.05 having
ciently enhanced fluidity extractions using different raw materials,
a significant influence on the extraction process for a confidence
such as aloe vera leaf skin, chamomile, grape seed or marigold [18].
level of 0.95. The sign associated with each of the effects indicates
Moreover, in some cases the modifier has little or no effect on the
a positive or negative influence on the yield caused by the variable
global yield but improves the extract composition with respect
in question.
to bioactive compounds. Therefore, the effect of pressure on the
Empirical correlations were obtained using the experimental
extraction process using co-solvents possibly depends on the type
data and the program Statgraphics Plus 5.1® . These correlations
of matrix and the interactions solute–matrix.
link the variables with influence on the global extraction yield of
The global extraction yields obtained using methanol or ethanol
the process using CO2 + 20% ethanol as solvent system.
as co-solvent showed no significant differences at a confidence level
Analysis of Table 2 and Fig. 3 indicates that the tempera-
of 0.95, notwithstanding the phenolic yields and the antioxidant
ture and the combined interaction of pressure/temperature are
activity of extracts showed remarkable differences. The extraction
the factors that influence the global extraction yield of the
yields of the main polyphenols of mango leaves (mangiferin and
process using CO2 + 20% ethanol. The combined interaction of pres-
quercetin 3-␤-d-glucoside) using CO2 plus co-solvents are shown
sure/temperature showed a negative effect on the extraction yield
in Table 1.
whereas the effect of temperature was positive. This explains the
The addition of methanol or ethanol helps the extraction of phe-
high extraction yield obtained at low pressure and high tempera-
nolic compounds because both co-solvents are capable of strong
ture.
interactions (hydrogen-bonding and dipole–dipole interactions)
The empirical correlation obtained for the global extraction
with phenols facilitating the extraction of these compounds. How-
yield is given below (Eq. (6))
ever, the highest quantities of polyphenols were recovered using
CO2 + 20% of ethanol showing this co-solvent is more efficient to Y = −8.6758 + 0.0304 · P + 0.290167 · T − 0.000726667 · P · T (6)
extract phenolic compounds from mango leaves. Y: global extraction yield (g/100 g of dry matter); T: temperature
The greater content on mangiferin and quercetin 3-␤-d- (◦ C); P: pressure (MPa). The resulting determination coefficient (R2 )
glucoside of mango leaves extracts obtained with CO2 + 20% of was 0.9961.
ethanol allows extracts to be obtained with better antioxidant
activities than CO2 + 20% of methanol extracts (Fig. 2B). The influ- 3.3. Subcritical Water Extraction
ence of P and/or T on the antioxidant activity of the extracts
obtained with CO2 + 20% (w/w) methanol/ethanol was analogous In spite of the strong antioxidant activity of the extracts obtained
to the effect described for the global yields. from mango leaves of the variety Osteen using CO2 + 20% of ethanol
Mango leaves extracts obtained at high pressures (40 MPa) using at 10 MPa and 55 ◦ C (1.23 ± 0.01 DPPH ␮g/␮g dry extract), such
methanol/ethanol as co-solvent presented a poor antioxidant activ- activity is still far from that AAI obtained for the antioxidant
ity with values of AAI lower than 0.5. By contrast, low pressures (+)-␣-tocopherol (3.84 DPPH ␮g/␮g dry extract). Moreover, the
(10 MPa) and high temperatures (55 ◦ C) led to obtained extracts extraction yields obtained with this solvent system were also not
with a strong AA (superior than 1.0) using ethanol as co-solvent, too high and are susceptible to be improved. In this way, SWE was
and a moderate AA using methanol. These high activities occurred explored using mango leaves as raw material so this technique is
because lower pressures enhanced the extraction of polyphe- considered highly efficient to extract polar or slightly polar com-
nols resulting in extracts with higher content on mangiferin and pounds from different natural matters [19–22].
quercetin and thus better antioxidant capacity. The global yield of the process obtained with SWE
Consequently, the extraction of antioxidant compounds from (24.24 ± 1.17%) from mango leaves of the variety Osteen was
mango leaves is more advantageous on working at low pressure significantly superior to those obtained with SC-CO2 and even
(10 MPa), high temperature (55 ◦ C) and with ethanol as CO2 modi- with CO2 plus co-solvents (see Fig. 2A). As well, the high efficiency
fier. Furthermore, ethanol provides an additional advantage as it is
considered to be a green solvent whose use in the pharmaceutical
Table 2
and food industries is not restricted. Estimated effects and analysis of variance for the solvent system CO2 + 20% ethanol.

Variable Effects p-Value

3.2.1. Analysis of the experimental design P −0.69 0.0384

Results obtained with CO2 + 20% of ethanol were statistically T 2.17 0.0041
PT −2.18 0.0041
analyzed by a multifactorial experimental design in order to ana-
lyze the influence of the variables on the process. P: pressure; T: temperature; PT: pressure/temperature.
 M.T. Fernández-Ponce et al. / J. of Supercritical Fluids 72 (2012) 168–175 173

0.02
0.01
0.02
0.07
0.03
0.01
0.01
(g/100 g
Content

extract)

±
±
±
±
±
±
±
0.36
0.64
1.23
0.88
1.69
1.56
0.98
3-␤-d-glucoside

6.9E−3

6.1E−3
2.3E−2
6.7E−3
3.8E−3
2.7E−4
5.0E−3
Yield (g/100 g
Rt = 34.4 min

dry leaves)
Quercetin

±
±
±
±
±
±
±
0.13
0.23
0.42
0.30
0.41
0.42
0.27
0.01
0.01
0.01
0.01
0.01
0.03
0.01
±
±
±
±
±
±
±
(g/100 g
Content

extract)

3.31
4.63
3.78
1.93
5.64
11.47
12.12
Fig. 4. Global yields obtained with CO2 + 20% ethanol (10 MPa, 55 ◦ C, 20 g/min of
CO2 , 3 h) and SWE (4 MPa, 100 ◦ C, 10 g/min, 3 h) from different mango cultivars: 1,
Kensington; 2, Kent; 3, Keitt; 4, Tommy Atkins; 5, Osteen; 6, Ataulfo; 7, Langra.

7.2E−4
2.5E−4
1.8E−3
1.4E−3
1.2E−3
9.1E−3
2.1E−3
Yield (g/100 g
Rt = 20.9 min
Mangiferin

dry leaves)
of subcritical water to extract phenolic compounds was also

±
±
±
±
±
±
±
demonstrated with the high recovery quantities of mangiferin and

1.10
1.64
1.29
0.65
1.37
3.09
3.37
quercetin (1365.9 ± 1.2 and 409.54 ± 6.7 mg/100 g of dry sample,

Antioxidant activity and phenolic content of extracts obtained with CO2 + 20% ethanol and Subcritical Water Extraction from different mango cultivars.
respectively) (see Table 1).
The high extraction of polyphenols with subcritical water can

(CDPPH /EC50 )

0.16

0.23
0.01
0.01
0.03
0.07

0.01
Antioxidant
be explained because of the fact that although water is a too polar
solvent not appropriate to extract polyphenols, its polarity can

activity

±
±
±
±
±
±
±
SWEb
be reduced with increasing temperature due to a reduction of its

6.61
6.76
4.20
4.02
7.92
6.50
5.94
AAI
hydrogen bonding propensity. In consequence, slightly polar com-

AAI, Antioxidant Activity Index expressed as DPPH ␮g/␮g dry extract was represented as the mean value ± standard deviations.
pounds, as polyphenols, with low solubility in water at ambient

CO2 + 20% ethanol was operated at best conditions obtained in the preliminary study (100 bar, 55 ◦ C, 20 g/min of CO2 and 3 h).

Yield for each phenolic compound expressed as g/100 g dry leaves was represented as the mean value ± standard deviations.
conditions or in SC-CO2 , can be much more soluble in subcritical

Content of phenolic compound expressed as g/100 g dry extract was represented as the mean value ± standard deviations.
0.02
0.02
0.04
0.01
0.05
0.01
0.06
water [19,20].

(g/100 g
Content

extract)
On the other hand, subcritical water extracts from mango

±
±
±
±
±
±
±
0.83
1.37
1.78
1.57
3.55
1.04
0.48
leaves of the variety Osteen present a potent antioxidant activ-
ity (7.92 ± 0.16 DPPH ␮g/␮g dry extract) (see Table 3) significantly
stronger than those obtained with CO2 plus ethanol. This activity
3-␤-d-glucoside

1.3E−3
1.2E−3
2.5E−3
2.7E−5
3.3E−3
6.6E−4
3.8E−3
Yield (g/100 g

was superior than (+)-␣-tocopherol, and similar to the antioxidant

Rt = 34.4 min

dry leaves)

activity of the standard compounds mangiferin (7.80 DPPH ␮g/␮g

Quercetin

dry extract) and quercetin (7.06 DPPH ␮g/␮g dry extract), showing
±
±
±
±
±
±
±
0.05
0.11
0.13
0.13
0.23
0.04
0.04
the high antioxidant capacity of subcritical water extracts.
In general, the antioxidant activity of extracts can be correlated
with their phenolic content. Nonetheless, the activity of each com-
pounds and the content of each one in the extract can influence on
0.01
0.01
0.01
0.01
0.01
0.01
0.01

the antioxidant activity too. So, the potent antioxidant capacity of

extract)e
(g/100 g
Content

±
±
±
±
±
±
±

the subcritical water extracts of the variety Osteen may be related

3.55
7.25
4.22
1.97
2.83
4.40
6.78

with their higher content on mangiferin than quercetin (Table 3),

but also the co-extraction of other antioxidant compounds con-
tributing in the activity of extracts.
9.1E−4
5.2E−4
9.7E−4
3.2E−4
4.2E−4
8.3E−4
4.6E−4
Yield (g/100 g
Rt = 20.9 min

The high efficiency of SWE to recover phenolic compounds and

dry leaves)d
Mangiferin

to obtain extracts with potent antioxidant activity shows that this

±
±
±
±
±
±
±

green extraction technology is a promising alternative to extract

0.23
0.58
0.30
0.16
0.18
0.19
0.50

SWE was operated at 40 bar, 100 ◦ C, 10 g/min and 3 h.

antioxidant compounds from mango leaves more efficiently than

CO2 + 20% ethanola

supercritical CO2 extraction.

(CDPPH /EC50 )

0.24
0.08
0.05
0.01
0.01
0.04
0.01

3.65 ± 0.01
7.80 ± 0.02
7.06 ± 0.01
Antioxidant

3.4. Evaluation of mango cultivars

activity

±
±
±
±
±
±
±
2.13
2.16
0.78
2.07
1.09
1.59
2.74
AAIc

The study of seven mango varieties (Kensington, Kent, Keitt,

Tommy Atkins, Osteen, Ataulfo and Langra) was carried out by
comparison of extraction yields, phenolic profile and antioxidant
Quercetin 3-␤-d-glucoside
Mangiferin ␤-d-glucoside

activity of the extracts obtained using CO2 + 20% ethanol at the best
Standard compounds

conditions obtained from the preliminary study (10 MPa, 55 ◦ C, a

(+)-␣-Tocopherol

flow rate of 20 g/min and 3 h), and also using subcritical water at
Mango cultivars

Tommy Atkins

100 ◦ C, 4 MPa, 10 g/min and 3 h.

Kensington

The global yields obtained with SWE from the seven varieties
Ataulfo

studied are shown in Fig. 4. SWE yields are higher than those
Osteen

Langra
Table 3

Keitt
Kent

obtained with CO2 + 20% ethanol for the entire mango cultivars
a

studied. The best global yields were obtained for the varieties Kent,
174 M.T. Fernández-Ponce et al. / J. of Supercritical Fluids 72 (2012) 168–175

Fig. 5. HPLC chromatograms of extracts obtained from Osteen variety using CO2 + 20% ethanol (A) and SWE (B): 1, mangiferin; 2, quercetin 3-␤-d-glucoside. Extraction with
CO2 + 20% ethanol was carried out at 10 MPa, 55 ◦ C, 20 g/min of CO2 , 3 h; and SWE at 4 MPa, 100 ◦ C, 10 g/min, 3 h.

Keitt, Tommy Atkins and Kensington (35.42 ± 0.53, 34.17 ± 0.73, the highest quercetin yields (0.41 ± 6.7E−3 to 0.42 ± 6.1E−3 g/100 g
33.71 ± 1.12 and 33.34 ± 2.04%, respectively) using SWE, while the dry leaves) while the lowest yield was for Kensington vari-
varieties Langra, Ataulfo and Osteen presented lower global yields ety (119.1 ± 0.1 mg/100 g dry matter). In relation to the yields
(27.84 ± 0.60, 26.96 ± 1.00 and 24.24 ± 1.17%, respectively). obtained with CO2 + 20% ethanol, the varieties Kent, Keitt, Tommy
When CO2 + 20% of ethanol was used as solvent system, the vari- Atkins and Osteen showed similar yields (0.11 ± 1.2E−3 to
eties Tommy Atkins and Kent presented the highest global yields 0.23 ± 3.3E−3 g/100 g dry leaves), and the lowest yields using this
(8.15 ± 0.31 and 8.06 ± 0.15%, respectively) followed by Langra, solvent system were obtained from Kensington, Ataulfo and Langra
Keitt, Kensington and Osteen (6.37 ± 0.18 to 7.37 ± 0.22%). How- varieties (0.04 ± 6.6E−4 to 0.05 ± 1.3E−3 g/100 g dry leaves).
ever, global yields obtained with CO2 extraction were inferior than The antioxidant activities and phenolic content of extracts
those obtained using subcritical water. obtained from the seven varieties are shown in Table 3. Extracts
The extraction yields of the main polyphenols present in mango obtained using CO2 + 20% of ethanol as solvent system showed good
leaves (mangiferin and quercetin 3-␤-d-glucoside) were analyzed antioxidant activities. Using this solvent system Langra extracts
by HPLC. The yields of the phenolic compounds in terms of mg/100 g presented the highest activity profile (2.74 ± 0.01 DPPH ␮g/␮g dry
dry matter for extracts obtained with CO2 + 20% ethanol and sub- extract) close to (+)-␣-tocopherol activity, and the cultivars Kens-
critical water are shown in Fig. 5, and typical HPLC chromatograms ington, Tommy Atkins, Kent, Kensington, Ataulfo and Osteen
of extracts obtained with both solvent systems are shown in Fig. 5. presented strong antioxidant activities with AAI values superior
The yields of the phenolic compounds mangiferin and quercetin than 1.0 ␮g/␮g dry extract.
3-␤-d-glucoside obtained with SWE were higher than those By contrast, subcritical water extracts showed a potent antioxi-
obtained with CO2 + 20% of ethanol (Table 3). For mangiferin, the dant activity up to (+)-␣-tocopherol activity (3.65 DPPH ␮g/␮g dry
varieties Ataulfo and Langra allowed high recoveries of this pheno- extract) and superior to CO2 + ethanol extracts even though in some
lic compound (3.37 ± 2.1E−3 and 3.09 ± 9.1E−3 g/100 g dry leaves) cases the content on mangiferin and quercetin 3-␤-d-glucoside of
using subcritical water as solvent system. These results were sim- extracts was similar than the phenolic content of CO2 + ethanol
ilar to the yield of mangiferin reported by other authors using extracts (see Table 3).
traditional solvent extraction methods from different mango leaves In this way, the potent antioxidant activity of subcritical water
varieties (3.71–9.36 g/100 g dry leaves) [23]. By contrast, the yields extracts (4.02–7.92 DPPH ␮g/␮g dry extract), similar to the pheno-
of mangiferin obtained with CO2 + 20% were inferior. The best val- lic compounds mangiferin and quercetin and significantly superior
ues were obtained for the varieties Kent and Langra (0.58 ± 5.2E−4 to CO2 + ethanol extracts, can be attributed to the high content
and 0.50 ± 4.6E−4 g/100 g dry leaves). Consequently, SWE would on mangiferin and quercetin, but also the presence of other polar
be a good alternative to traditional methods to extract phenolic antioxidant compounds, not identified in this work, which also
compounds from mango leaves. contribute to their antioxidant activity.
About the extraction of quercetin 3-␤-d-glucoside, Table 3
shows that the recovery of this phenolic compound was inferior 4. Conclusions
to mangiferin. However, SWE has proved to be also more effi-
cient than CO2 + 20% ethanol to extract this polyphenol. Subcritical The results presented in this study show that SWE is an excellent
water extracts from varieties Ataulfo, Osteen and Keitt showed alternative to recover high quantities of phenolic compounds from
 M.T. Fernández-Ponce et al. / J. of Supercritical Fluids 72 (2012) 168–175 175

mango leaves thanks to the high extraction yields and the potent [16] R.N. Carvalho Jr., L.S. Moura, P.T.V. Rosa, M.A. Meireles, Supercritical fluid
antioxidant activity of extracts obtained by this extraction method. extraction from rosemary (Rosmarinus officinalis): kinetic data, extract’s global
yield, composition, and antioxidant activity, J. Supercritical Fluids 35 (2005)
SWE resulted more efficient than SC-CO2 extraction even though 197–204.
the performance of CO2 extraction was improved with the addi- [17] P.F. Leal, M.B. Kfouri, F.C. Alexandre, F.H.R. Fagundes, J.M. Prado, M.H. Toyama,
tion of ethanol as CO2 modifier. Quality of subcritical water extracts M.A. Meireles, Brazilian Ginseng extraction via LPSE and SFE: global yields,
extraction kinetics, chemical composition and antioxidant activity, J. Super-
was significantly superior to that obtained with the subcritical mix- critical Fluids 54 (2010) 38–45.
tures of CO2 + 20% of ethanol in terms of phenolic content and also [18] B. Díaz-Reinoso, A. Moure, H. Domínguez, J.C. Parajo, Supercritical CO2 extrac-
antioxidant activity. Subcritical water extracts from the variety tion and purification of compounds with antioxidant activity, J. Agricultural
and Food Chemistry 54 (2006) 2441–2469.
Osteen presented a potent antioxidant activity; nevertheless, the
[19] W. Kim, B. Veriansyah, Y. Lee, J. Kim, J. Kim, Extraction of mangiferin from
other varieties studied also presented a good performance show- Mahkota Dewa (Phaleria macrocarpa) using subcritical water, J. Industrial and
ing mango leaves are an attractive raw material to obtain products Engineering Chemistry 16 (2010) 425–430.
[20] C. Turner, P. Turner, G. Jacobson, K. Almgren, M. Waldebäck, P. Sjöberg,
with potential applications in food, cosmetic and pharmaceutical
E. Nordberg Karlsson, k. E. Markides, Subcritical water extraction and b-
industries. Optimization of SWE to recover phenolic compounds glucosidase-catalyzed hydrolysis of quercetin glycosides in onion waste, Green
from mango leaves will be realized in future works having into Chemistry 8 (2006) 949–959.
account the high efficiency showed for this extraction technique. [21] M. Ko, C. Cheigh, S. Cho, M. Chung, Subcritical water extraction of flavonol
quercetin from onion skin, J. Food Engineering 102 (2011) 327–333.
[22] Z.Y. Ju, Subcritical water and sulfured water extraction of anthocyanins
Acknowledgments and other phenolics from dried red grape skin, J. Food Sciences 70 (2005)
S270–S276.
[23] J.C. Barreto, M.T.S. Trevisan, W.E. Hull, G. Erben, E.S. De Brito, B. Pfundstein,
The authors thank Science and Innovation secretaryship of G. Würtele, B. Spiegelhalder, R.W. Owen, Characterization and quantitation of
the Spanish Government for financial support (Project CTQ2011- polyphenolic compounds in bark, kernel, leaves, and peel of mango (Mangifera
22974), which enabled this work to be carried out, and to the indica L.), J. Agricultural and Food Chemistry 56 (2008) 5599–5610.
[24] M. Massibo, Q. He, Major mango polyphenols and their potential significance
research group of “Estación Experimental La Mayora” (CSIC, Málaga, to human health, Comprehensive Reviews in Food Science and Food Safety 7
Spain) for providing the raw material, especially to Dr. Jorge Gon- (2008) 309–319.
zalez Fernández and Dr. Emilio Guirado for their support. [25] N. Berardini, R. Fezer, J. Conrad, U. Beifuss, R. Carle, A. Schieber, Screening
of mango (Mangifera indica L.) cultivars for their contents of flavonol O- and
xanthone C-glycosides, anthocyanins, and pectin, J. Agricultural and Food
References Chemistry 53 (2005) 1563–1570.
[26] R.M.Y. Morsi, N.R. El-Tahan, A. del M.A. El-Hadad, Effect of aqueous extract
[1] B. Kamm, M. Kamm, Principles of biorefineries, Applied Microbiology and Mangifera indica leaves, as functional foods, J. Applied Sciences Research 6
Biotechnology 64 (2004) 137–145. (2010) 712–721.
[2] J. Clark, V. Budarin, F. Deswarte, J. Hard, Kerton, A. Hunt, R. Luque, D. Macquarrie, [27] G. Noratto, M. Bertoldi, K. Krenek, S. Talcott, P.C. Stringheta, S.U.
L. Milkowski, A. Rodriguez, O. Samuel, S. Tavener, R. White, A. Wilson, Green Mertens-Talcott, Anticarcinogenic effects of polyphenolics from mango
chemistry and the biorefinery: a partnership for a sustainable future, Green (Mangifera indica) varieties, J. Agricultural and Food Chemistry 58 (2010)
Chemistry 8 (2006) 853–860. 4101–4112.
[3] J. Clark, Perspective green chemistry for the second generation [28] C.M. Ajila, K.A. Naidu, s.G. Bhat, U.J.S. Prasada Rao, Bioactive compounds
biorefinery—sustainable chemical manufacturing based on biomass, J. and antioxidant potential of mango peel extract, Food Chemistry 105 (2007)
Chemical Technology and Biotechnology 82 (2007) 603–609. 982–988.
[4] M.R. Maróstica, A. Leite, N. Romanelli, V. Dragano, Supercritical fluid extraction [29] J. Rodriguez, D. Di Pierro, M. Gioia, S. Monaco, R. Delgado, M. Coletta, S. Marini,
and stabilization of phenolic compounds from natural sources—review (super- Effects of a natural extract from Mangifera indica L, and its active compound,
critical extraction and stabilization of phenolic compounds), Open Chemical mangiferin, on energy state and lipid peroxidation of red blood cells, Biochimica
Engineering J. 4 (2010) 51–60. et Biophysica Acta 1760 (2006) 133–134.
[5] M.A.A. Miereles, Supercritical fluid extraction of medicinal plants, Electronical [30] S.M. Rocha Ribeiro, J.H. Queiroz, M.E. Lopes Ribeiro, F.C. Campo, H.M. Piheiro,
J. Environmental, Agricultural and Food Chemistry 7 (2008) 3254–3258. Antioxidant in mango (Mangifera indica L.) pulp, Plant Food Human Nutrition
[6] H. Castro-Vargas, L. Rodríguez-Varela, S. Ferreira, F. Parada-Alfonso, Extraction 62 (2007) 13–17.
of phenolic fraction from guava seeds (Psidium guajava L.) using supercritical [31] FAO, April 2012. http://faostat.fao.org/
carbon dioxide and co-solvents, J. Supercritical Fluids 51 (2010) 319–324. [32] C. Pereira, M.A. Meireles, Evaluation of global yield, composition, antioxidant
[7] C. Mantell, M. Rodríguez, E. Martínez de la Ossa, A screening analysis of the activity and cost of manufacturing of extracts from lemon verbena (Aloysia
high-pressure extraction of anthocyanins from red grape pomaces with carbon triphylla [L’Hérit.] britton) and Mango (Mangifera indica L.) leaves, J. Food Process
dioxide and cosolvent, Engineering in Life Sciences 3 (2003) 38–42. Engineering 30 (2007) 150–173.
[8] L. Casas, C. Mantell, M. Rodríguez, E. Martínez de la Ossa, A. Roldán, L. De Ory, I. [33] W. Brand-Williams, M.E. Cuvelier, C. Berset, Use of a free radical method to
Caro, A. Blandino, Extraction of resveratrol from the pomace of Palomino fino evaluate antioxidant activity, Lebensmittel Wissenschaft und Technologie 28
grapes by supercritical carbon dioxide, J. Food Engineering 96 (2010) 304–308. (1995) 25–30.
[9] Q. Hu, Y. Hu, J. Xu, Free radical-scavenging activity of Aloe vera (Aloe barbadensis [34] R. Scherer, H. Teixeira Godoy, Antioxidant activity index (AAI) by the 2,2-
Miller) extracts by supercritical carbon dioxide extraction, Food Chemistry 91 diphenyl-1-picrylhydrazyl method, Food Chemistry 112 (2009) 654–658.
(2005) 85–90. [35] P. Gerbaud, La Mangue, Le Dossiers de FruiTrop 164 (2009) 7–39.
[10] K.G. Martino, D. Guyer, Supercritical fluid extraction of quercetin from onion [36] V. Galán Sauco, El cultivo del mango, 2nd ed., Ediciones Mundi-Prensa, Madrid,
skins, J. Food Process Engineering 27 (2004) 17–28. 2009, pp. 50–150.
[11] E. Dimitrieska-Stojkovic, Z. Zdravkovski, Supercritical fluid extraction of [37] K.S. Andrade, R.T. Gonclvez, M. Maraschin, R.M. Ribeiro-do-Valle, J. Martinez,
quercetin and rutin from Hyperici Herba, J. Liquid Chromatography Related S.R.S. Ferreira, Supercritical fluid extraction from spent coffee grounds and cof-
Technologies 26 (2003) 2517–2533. fee husks: antioxidant activity and effect of operational variables on extract
[12] A.S. Zarena, K. Udaya Sankar, Supercritical carbon dioxide extraction of xan- composition, Talanta 88 (2012) 544–552.
thones with antioxidant activity from Garcinia mangostana: characterization [38] A. Chafer, T. Fornari, A. Berna, R.P. Stateva, Solubility of quercetin in supercritical
by HPLC/LC–ESI-MS, J. Supercritical Fluids 49 (2009) 330–337. CO2 + ethanol as a modifier: measurements and thermodynamic modelling, J.
[13] M. Herrero, A. Cifuentes, W. Ibañez, Sub- and supercritical fluid extraction Supercritical Fluids 32 (2004) 89–96.
of functional ingredients from different natural sources: plants, food-by- [39] I.H. Adil, H.I. Cetin, M.E. Yener, A. Bayindirli, Subcritical (carbon diox-
products, algae and microalgae. A review, Food Chemistry 98 (2006) 136–148. ide + ethanol) extraction of polyphenols from apple and peach pomaces, and
[14] T. Vatai, M. Skerget, Z. Knez, Extraction of phenolic compounds from elder berry determination of the antioxidant activities of the extracts, J. Supercritical Fluids
and different grape marc varieties using organic solvents and/or supercritical 47 (2007) 55–63.
carbon dioxide, J. Food Engineering 90 (2009) 246–254. [40] M. Palma, L.T. Taylor, Extraction of polyphenolic compounds from grape
[15] L.E. Laroze, B. Díaz-Reinoso, A. Moure, M.E. Zúñiga, H. Domínguez, Extraction seeds with near critical carbon dioxide, J. Chromatography A 879 (1999)
of antioxidants from several berries pressing wastes using conventional and 117–124.
supercritical solvents, European Food Research and Technology 231 (2010) [41] L. Wang, C.L. Weller, Recent advances in extraction of nutraceuticals from
669–677. plants, Trends in Food Science & Technology 17 (2006) 300–312.

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.