C

:
F
o
o
d
C
h
e
m
i
s
t
r
y
Investigation on the Antioxidant Activity
of Leaves, Peels, Stems Bark, and Kernel
of Mango (Mangifera indica L.)
Bushra Sultana, Zaib Hussain, Muhammad Asif, and Adil Munir
Abstract: Bioactive polyphenols, cartenoids, and anthocyanins present in fruits and vegetables are receiving much
attention because of their potential antioxidant activity. This study was conducted to determine antioxidant activity of
leaves, peels, stem bark, and kernel of mango varieties langra and chonsa. Total phenolic (TPC) and total avonoid
contents (TFCs) in segments of langra ranged from 63.89 to 116.80 mg GAE/g DW and 45.56 to 90.89 mg CE/g DW,
respectively, and that of chonsa were 69.24 to 122.60 mg GAE/g DW and 48.43 to 92.55 mg CE/g DW, respectively.
The 2, 2-diphenyl-1-picrylhydrazyl (DPPH) scavenging activity and linoleic inhibition capacity in segments of langra
ranged from 53.30% to 61.10% and 40.0% to 47.20%, respectively, whereas for chonsa; 56.40% to 66.0% and 48.1% to
49.0%, respectively. The reducing potentials of different segments of langra and chonsa at concentration of 10 mg/mL
were 0.512 to 0.850 and 0.595 to 0.665 mV, respectively. Comparison between both varieties showed chonsa exhibited
better antioxidant activity. Data were analyzed by analysis of variance (ANOVA) using completely randomised design
(CRD) under factorial.
Keywords: antioxidant, DPPH, mango, oxidation, TFC, TPC
Introduction
Antioxidants are used to preserve and protect foods from ran-
cidity, discoloration, or deterioration caused by autoxidation and
are commonly used to improve the shelf life and stability of lipids
and lipid-containing foods.
Numerous epidemiological studies suggest that diets rich in
phytochemicals and antioxidants execute a protective role in health
improvement and disease prevention. Frequent consumption of
fruits and vegetables is associated with a subordinate risk of cancer,
heart disease, hypertension and stroke (Lako and others 2007).
Fruits, vegetables and whole grain foods provide protection which
can slow down the process of oxidative damage. They contain
a variety of natural antioxidants and are considered to be more
benecial than antioxidant supplements (Ajila and others 2007).
Recent studies have shown that many avonoids and related
polyphenols contribute signicantly to the total antioxidant ac-
tivity of many fruits and vegetables (Torunn and others 2007;
Othman and others 2009; Babbar and others 2011). Interest in
natural antioxidants has led to the investigation of antioxidants in
fruits, vegetables, herbs, spices, cereals and agrowastes (Kuan and
others 2011). Special attention has been paid to fruits, as they are
rich in phenolics, avonoids and vitamins (Jeong and others 2004;
Calder on and others 2011).
Mango (Mangifera indica L.), belonging to the family Anacar-
diaceae and order Rutales, is one of the most popular, widely con-
sumed edible fruits in the world and its global production ranks 5th
MS 20111405 Submitted 11/21/2011, Accepted 5/14/2012. Authors Sultana,
Asif, and Munir are with Dept. of Chemistry & Biochemistry, Univ. of Agricul-
ture, Faisalabad, Pakistan. Author Hussain is with Inst. of Chemistry, Univ. of the
Punjab, Lahore, Pakistan. Direct inquiries to author Hussain (E-mail: drzh1972@
hotmail.com).
among major fruit crops. Mangoes are grown in various regions,
especially in the tropics. Mango peels and kernel are generated
as by-products during juice and canned mango manufacturing.
Mango peels and kernel have been reported to be a rich source of
gallates, gallotannins, xanthone glucosides, avonols, ascorbic acid,
carotenoids, enzymes and dietary ber (Ajila and others 2007).
Mango like other yellow/orange fruit, such as pumpkin and car-
rot, is an excellent source of beta-carotene (vitamin A). It is also
rich in vitamin C with traces of vitamins E, B and K (Kim and oth-
ers 2009). Mangoes also contain potassium, making them ideal for
hypertensive patients or those who want to replenish energy after
physical activity. They are also high in antioxidants and low in car-
bohydrates (15% sugars) (Ajila and others 2007). The nutritional
value of mango may contribute to weight gain but it can help to
reduce disorders such as, hair loss, heart stroke, prickly heat, bac-
terial infections, sinusitis, piles, indigestion, constipation, morning
sickness, diarrhea, dysentery, scurvy, spleen enlargement, liver dis-
orders, menstrual disorders, leucorrhea and virginities (Atawodi,
2005). Several studies have addressed the levels of antioxidants in
various segments of the fruit but few have compared and identi-
ed which component contains the highest level of antioxidants
(Einbond and others 2004).
To the best of our knowledge, no literature report is available
on antioxidant activity of different segments of mango. Therefore,
the present research work has been designed in order to evaluate
the antioxidant activity of the different segments of mango, that
is, peels, leaves, stem bark and kernel of 2 mango varieties (lungra
and chonsa) which are commonly consumed in Pakistan.
Materials and Methods
Collection of samples
Fresh samples of peels, kernel, leaves and stem bark of 2 varieties
(chonsa and langra) of mango (Mangifera indica L.) were collected
C
2012 Institute of Food Technologists
R
doi: 10.1111/j.1750-3841.2012.02807.x Vol. 77, Nr. 8, 2012
r
Journal of Food Science C849
Further reproduction without permission is prohibited
C
:
F
o
o
d
C
h
e
m
i
s
t
r
y
Investigation on the antioxidant activity . . .
from a mango garden situated in the nearby area of Faisalabad,
Pakistan. The samples were further identied and authenticated
from Dept. of Botany, Univ. of Agriculture, Faisalabad.
Pretreatment of samples
The peels and kernel were separated from the mango esh
by using a sharp steel knife. Mango tree leaves and stem bark
were washed with distilled water. The samples were then dried at
room temperature, ground into ne powder, and subjected to an
extraction procedure.
Extraction
The ground mango (Mangifera indica L.) samples (10 g) were ex-
tracted separately with 100 mL of 80% methanol (methanol:water,
80:20, v/v) and agitated for 24 h at room temperature in an orbital
shaker (Gallenkamp, Loughborough, Leicestershire, UK) under
light conditions at a speed of 150 rpm. All extracts were sepa-
rated from the residues by ltration using Whatman No. 1 lter
paper. The residues were extracted twice in the same manner and
the extracts combined. The combined extracts were concentrated
under reduced pressure at 45

C, using a rotary evaporator. The
dried crude extracts were weighed to calculate the yield and sub-
sequently stored at 4

C for further analysis.
Determination of total phenol (TP) concentration
The amount of TP was assessed using the FolinCiocalteu
procedure as described by Chaovanalikit and Wrolstad (2004).
Dry mass of each extract (50 mg) was mixed with 0.5 mL of
FolinCiocalteu reagent and 7.5 mL deionized water. The mix-
ture was kept at room temperature for 10 min to which 1.5 mL of
20% (w/v) sodium carbonate was added. The mixture was heated
in a water bath at 40

C for 20 min and then cooled in an ice
bath. Finally, the absorbance at 755 nm was measured using a spec-
trophotometer (Hitachi U-2001, model 121-0032, Tokyo, Japan).
The results were expressed as gallic acid equivalents (GAEs) per
dry matter. All samples were analyzed in triplicate and the results
averaged.
Determination of total avonoid (TF) concentration
TF concentrations were determined following the procedure
described by Dewanto and others (2002). One milliliter of aque-
ous extract containing 0.1 g/mL of extract was placed in a 10-mL
volumetric ask to which 5 mL of distilled water was added fol-
lowed by 0.3 mL of 5% NaNO
2
. After 5 min, 0.6 mL of 10%
AlCl
3
was added and after a further 5 min, 2 mL of 1 M NaOH
was added and the volume made up with distilled water. The so-
lution was mixed and absorbance was measured at 510 nm. TF
amounts were expressed as catechin equivalents. All samples were
analyzed in triplicate and the results averaged.
Antioxidant activity determination in linoleic acid system
The antioxidant activity of extracts was determined in terms
of measurement of percentage (%) inhibition of peroxidation in
linoleic acid system following a reported method of Iqbal and
Bhanger (2006). Extracts (5 mg) of each sample were added to a
mixture of linoleic acid (0.13 mL), 99.8% ethanol (10 mL) and
0.2 M phosphate buffer (pH 7, 10 mL). The total mixture was
diluted to 25 mL with distilled water. The solution was then
incubated at 40

C and the degree of oxidation was measured
following the thiocyanate method whereby 10 mL of ethanol
(75%), 0.2 mL of an aqueous solution of ammonium thiocyanate
(30%), 0.2 mL of sample solution and 0.2 mL of ferrous chloride
(FeCl
2
) solution (20 mM in 3.5% HCl) were added sequentially.
After 3 min of stirring, the absorption value of the mixtures was
measured at 500 nm (peroxide contents). A control experiment
was performed with linoleic acid with omission of the extracts.
Butylated hydroxytoluene (BHT) and ascorbic acid (200 ppm)
were used as positive controls. The maximum peroxidation level
was observed as 360 h (15 d) in the sample which contained
no antioxidant component and was used as a test point. Percent
inhibition of linoleic acid peroxidation was calculated using the
following formula:
I % = 100

Absorbance increase of control at 360 h

Absorbance increase of sample at 360 h

Absorbance increase of control at 360 h

100
(1)
Determination of reducing power
The reducing power of the extracts was determined accord-
ing to the procedure described by Yen and others (2000), with
modication. Equivalent volume of extracts containing 2 to 10.0
mg of extracts was mixed with sodium phosphate buffer (5.0 mL,
0.2 M, pH 6.6) and potassium ferricyanide (5.0 mL, 1.0%). The
mixture was incubated at 50

C for 20 min after which 5 mL of
10% trichloroacetic acid was added and centrifuged for 10 min
at 5

C in a refrigerated centrifuge (CHM-17; Kokusan Denki,
Tokyo, Japan). The upper layer of the solution (5.0 mL) was di-
luted with 5.0 mL of distilled water and ferric chloride (1.0 mL,
0.1%). The absorbance was determined at 700 nm. The analysis
was performed in triplicate and the results averaged.
2, 2-Diphenyl-1-picrylhydrazyl (DPPH) radical scavenging
assay
Free radical scavenging activities of the extract were measured
by using the procedure described by (Iqbal and Bhanger (2006).
To 1.0 mL of each extract containing 0.025 mg/mL of extract in
methanol, 5.0 mL of 0.025 g/L freshly prepared solution of DPPH
was added. Absorbance at 0, 0.5, 1, 2, 5, and 10 min was measured
at 515 nm and the remaining amounts of DPPH free radical were
calculated from the calibration curve. Absorbance measured after
5 min was used to compare the radical scavenging activity of each
extract.
Statistical analysis
Each sample was extracted in triplicate and each extract was
analyzed in triplicate. The descriptive statistics (mean SD) were
worked out. Minitab 2000 Version 13.2 (Minitab Inc., Pa., USA)
was used for the analysis of variance, t-test and comparison of
means.
Results and Discussion
Yield (%), total phenolic (TPC), and total avonoid
contents (TFC)
The percentage yield of the extracts from leaves, peels, stems
bark, and kernel of langra variety was found to be 26.6%, 27.8%,
24.7% and 25.5% respectively, while for chonsa variety, it was
27.1%, 28.3%, 25.2% and 25.7% respectively. However, for each
C850 Journal of Food Science
r
Vol. 77, Nr. 8, 2012
C
:
F
o
o
d
C
h
e
m
i
s
t
r
y
Investigation on the antioxidant activity . . .
variety studied, all these differences were statistically non signi-
cant (Table 1). On an average, the maximum yield (28.3%) was
obtained from chonsa peels while the minimum (24.7%) from
langra stem bark. Results showed that all the segments of chonsa
variety showed a slight increase in percentage yield as compared
to langra variety. The values of percentage yield determined in the
present investigation were higher than those previously reported
in mango (Mangifera indica L.) (Ling and others 2009). Differences
in the yield of extracts can be attributed to the availability of
different extractable components, maturity, nature of variety and
agroclimatic conditions (Hsu and others 2006).
Recently, phenolics and avonoids have acquired considerable
interest because of their potential benecial effects on human
health. They have been reported to show antiviral, anti-allergic,
antiplatelet, anti-inammatory, anticancer, and antioxidant activi-
ties. Total phenolic content in extracts of langra and chonsa mango
(Mangifera indica L.) varieties were determined by Folin-Ciocalteau
method. This method was chosen due to its sensitivity, low inter-
ference and fastness to quantify the phenolic contents (Sultana and
others 2007). The TPC and TFC of the 2 varieties of mango are
given in Table 1.
The TPCand TFCcontents of different segments, that is, leaves,
peels, stems bark and kernel of mango fruit of langra variety were
found to be 86.62, 116.80, 78.56, 63.89 mg GAE/g and 66.54,
90.89, 56.87, 45.56 mg CE /g of dry weight, respectively. Almost,
all these differences were signicant (P < 0.05). The TPC and
TFC contents of different segments of chonsa variety were found
to be 93.18, 122.60, 78.56, 69.24 mg GAE/g and 83.67, 92.55,
Table 1Yield (%) TPC and TFC of different segments of langra
and chonsa mangoes.
Yield (%)
a
TPC (mg/g)
b
TFC (mg/g)
c
Mango langra
Leaves 26.6 1.77a 86.62 0.12c 76.54 0.15c
Peels 27.8 0.56a 116.80 0.98d 90.89 1.06d
Kernel 25.5 2.71a 63.89 0.72a 45.56 0.38a
Stem bark 24.7 0.95a 75.95 0.84b 56.87 0.45b
Mango chonsa
Leaves 27.1 1.22a 93.18 1.20c 83.67 0.38c
Peels 28.3 0.23a 122.60 0.56d 92.55 1.27c
Kernel 25.7 0.85a 69.24 0.54a 48.43 0.21a
Stem bark 25.2 1.10a 78.56 1.34b 65.45 0.12b
Means within a column followed by different letters for any 1 variety of mango are
signicantly different at P < 0.05.
a
Percent yield of dry extract.
b
Total phenolic content.
c
Total avonoid content.
Figure 1Percentage of inhibition of linoleic acid of different segments of
2 varieties of mango. BHT, butylated hydroxytoluene. Values are mean of
triplicate determination.
65.45, 48.43 mg CE /g of dry weight, respectively. The maximum
total phenolic contents (122.60 mg GAE/g DW) were obtained
from chonsa peels while the minimum (63.89 mg GAE/g DW)
were obtained from langra stem bark. The values of total phenolic
contents determined in mango kernel in the present investigation
were lower than earlier reported values (117 mg GAE/g DW)
(Soong and Barlow 2004). However, the values of total phenolic
contents determined in the present investigation were comparable
to those reported in mango peel (54.67 to 109.76 mg GAE/g)
(Ajila and others 2007).
The maximum value of avonoids, 92.55 mg CE/g of dry
weight, was obtained in chonsa peels while the minimum, 45.56
mg CE /g of dry weight, was observed in langra kernel extract.
Comparison of the 2 different varieties of mango fruit revealed
that the different segments of chonsa variety exhibited better TPC
and TFC values.
Antioxidant activity in linoleic acid system
Higher absorbance indicates higher concentration and hence
lower antioxidant activity. Considerable inhibition of peroxidation
may be attributed to the presence of established antioxidants, such
as xanthones, avans, avonols and di-anthraquinones along with
other phenolic compounds in the extracts (Sultana and others
2007).
The percentage inhibition of linoleic acid obtained in the
2 varieties of mango is shown in Figure 1. The percentage in-
hibition shown in different segments of langra variety was found
to be 57.48%, 62.23%, 47.38% and 24.85% respectively and for
chonsa variety 60.9%, 64.63%, 26.82% and 49.84% respectively.
The levels of percentage inhibition in the different segments of
langra and chonsa were compared with BHT which were 79.67%
and 75.87%. Percentage of inhibition of linoleic acid by BHT was
signicantly (P < 0.05) higher as compared to various parts of
2 varieties of mangoes. Comparison of percentage inhibition of
different segments of the 2 varieties depicted that chonsa peels had
maximum inhibition capacity 64.63% while the langra stem bark
had minimum inhibition capacity (24.85%). However, the effect
of varieties had a non-signicant effect. The values of inhibition
of oxidation determined in the present investigation were higher
than those reported in the literature (Kim and others 2010).
DPPH radical scavenging activity
To evaluate free radical scavenging activity, the extracts were
allowed to react with a stable free radical, DPPH. The DPPH
Figure 2DPPH radical scavenging activity of different segments of 2 va-
rieties of mango. BHT, butylated hydroxytoluene; BHA, butylated hydrox-
yanisole. Values are mean of triplicate determination.
Vol. 77, Nr. 8, 2012
r
Journal of Food Science C851
C
:
F
o
o
d
C
h
e
m
i
s
t
r
y
Investigation on the antioxidant activity . . .
Table 2Reducing power (mV) of extracts of 4 segments of Langra and Chonsa mangoes.
Mango (langra) Mango (chonsa)
Concentration (mg/mL) Concentration (mg/mL)
2 5 7 10 2 5 7 10
Leaves 0.212 .03a 0.602 .05bc 0.723 .11c 0.512 .07b 0.413 .04a 0.487 .07b 0.523 .15b 0.637 .08c
Peels 0.234 .02a 0.292 .13a 0.372 .04b 0.850 .21c 0.447 .04a 0.513 .12ab 0.567 .16bc 0.624 .07c
Kernel 0.423 .02a 0.483 .09ab 0.545 .04bc 0.607 .12c 0.52 0.03b 0.448 .05a 0.510 .20b 0.595 .08c
Stem bark 0.534 .03a 0.611 .05ab 0.692 .04bc 0.741 .12c 0.571 .05a 0.559 .03a 0.605 .14b 0.665 .06c
Means within a row for any 1 variety of mangoes, followed by different letters are signicantly different at P < 0.05 level.
radical, which is of deep violet color, gives intensive absorption
within the range of 515 to 517 nm.
Results obtained from the present analysis are presented in
Figure 2 and show that the scavenging activities of different
segments of langra variety were 53.3%, 61.1%, 47.2% and 40.0%
respectively, while for chonsa variety the values were found to be
56.4%, 66.0%, 49.0% and 48.1.% respectively. These values were
compared with the values of BHT and BHA which were 72.2%
and 70.3% respectively. Percentage of DPPH radical scavenging
activity was highest in case of BHT (72.2%) followed by BHA
(70.3%). These differences were non-signicant statistically. How-
ever, DPPH radical scavenging activity of BHT was signicantly
(P < 0.05) higher when compared with various parts of 2 varieties
of mangoes. The highest scavenging activity was achieved with
chonsa peels (66.0%) extract while the lowest with langra stem
bark extract (40.0%). Comparison of the 2 varieties of mango
fruit revealed that chonsa exhibited better free radical scavenging
activity as compared to langra. The effect of variety had a non-
signicant effect on DPPH radical scavenging activity of various
parts of mango. The values of free radical scavenging activity de-
termined in the present investigation were comparable to those
reported in apple (14.81% to 71.89%) (Peschel and others 2006).
Reducing power
The measurement of reducing power can reect several aspects
of antioxidant activity in the sample. In this method ferric ions
are reduced to ferrous ions resulting in a change in color from
yellow to bluish green. The intensity of color depends on reduc-
ing potential of the compounds present in the medium. Greater
the intensity of the color results in greater absorption, hence, an
increase in antioxidant activity.
The reducing power of 4 segments of chonsa and langra ex-
tracts (Table 2) was found to be in the range of 0.512 to 0.850 mV.
Chonsa exhibited reducing power in the range 0.595 to 0.665 mV
at a concentration of 10 mg/mL. The values of reducing power
determined in the present investigation were lower than those
reported in chestnut fruit (Barreira and others 2008). Statistical
analysis showed the signicant (P < 0.05) differences regarding
reducing power between the fruit varieties and also among differ-
ent segments (Table 2).
Conclusion
The study demonstrated that different segments of chonsa had
higher antioxidant capacity while the counterparts of langra dis-
played lower antioxidant capacity. The comparison among dif-
ferent segments showed that chonsa stem bark exhibited higher
TFC, TPC, and reducing power. On an average, Langra peel
showed a marked difference in reducing power for extracts 7 to 10
(Table 2). It can be concluded that chonsa peel is a benecial
by-product of the mango processing industry and a rich source of
bioactive compounds.
Acknowledgment
The authors wish to thank Manzar Islam (Univ. of the Punjab,
Lahore) for his supportive technical assistance and guidance.
References
Ajila CM, Naidu KA, Bhat SG, Rao P. 2007. Bioactive compounds and antioxidant potential
of mango peel extract. Food Chem 105:9828.
Ajila CM, Bhat SG, Rao P. 2007. Valuable components of raw and ripe peels from two Indian
mango varieties. Food Chem 102:100611.
Atawodi SE. 2005. Antioxidant potential of African medicinal plants. Afr J Biotechnol
4(2):12833.
Barreira JCM, Ferreira ICFR., Oliveira MBPP, Pereira JA. 2008. Antioxidant activities of the
extracts from chestnut ower, leaf, skins and fruit. Food Chem 107:110611.
Babbar N, Oberoi HS, Uppal DS, Patil RT. 2011. Total phenolic content and antioxidant
capacity of extracts obtained from six important fruit residues. Food Res Int 44:3916.
Calder on JC, Jaimes LC, Hern andez EG, Villanova BG. 2011. Antioxidant capacity, phenolic
content and vitamin C in pulp, peel and seed from 24 exotic fruits from Colombia. Food Res
Int 44:204753.
Chaovanalikit A, Wrolstad RE. 2004. Total anthocyanins and total phenolics of fresh and pro-
cessed cherries and their antioxidant properties. Food Sci 69(1):6772.
Dewanto V, Wu X, Adom KK, Liu RH. 2002. Thermal processing enhances the nutritional
value of tomatoes by increasing total antioxidant activity. Food Chem 50:30104.
Einbond LS, Reynertson KA, DongLuo X, Basile MJ, Kennelly EJ. 2004. Anthocyanin antiox-
idants from edible fruits. Food Chem 84:238.
Hsu B, Coupar IM, Ng K. 2006. Antioxidant activity of hot water extract from the fruit of the
Doum palm, Hyphaene thebaica. Food Chem 98:31728.
Iqbal S, Bhanger MI. 2006. Effect of season and production location on antioxidant activity of
Moringa oleifera leaves grown in Pakistan. J Food Comp Anal 19:54451.
Jeong SM, Kuo SYD, Kim RS, Jo C, Nam KC, Ahn DU, Lee SC. 2004. Effect of heat treatment
on the antioxidant activity of extracts from citrus peels. J Agric Food Chem 52:338993.
Kim Y, Singleton AJL, Talcott ST. 2009. Antioxidant phytochemical and quality changes as-
sociated with hot water immersion treatment of mangoes (Mangifera indica L.). Food Chem
115:98993.
Kim H, Moon JY, Kim H, Lee D, Cho M, Choi H, Kim Y, Mosaddik A, Cho KS. 2010.
Antioxidant and antiproliferative activities of mango (Mangifera indica L.) esh and peel. Food
Chem 121:42936.
Kuan C, Yuen K, Bhat R, Liong M. 2011. Physicochemical characterization of alkali treated
fractions from corncob and wheat straw and the production of nanobres. Food Res Int
44:28229.
Lako J, Trenerry VC, Wahlqvist M, Wattanapenpaiboon N, Sotheeswaran S, Premier R. 2007.
Phytochemical avonols, carotenoids and the antioxidant properties of a wide selection of
Fijian fruit, vegetables and other readily available foods. Food Chem 101:172741.
Ling LT, Yap SA, Radhakrishnan AK, Subramaniam T, Cheng HM, Palanisamy UD. 2009.
Standardised Mangifera indica extract is an ideal antioxidant. Food Chem 113:115459.
Othman A, Ismail A, Nawalyah AG, Ilham A. 2009. Total antioxidant activity in 35 Ugandan
fruits and vegetables. Food Chem 113:8591.
Peschel W, Rabaneda FS, Diekmann W, Plescher A, Gartz I, Jime D, Ravento RL, Buxaderas
S, Codina C. 2006. An industrial approach in the search of natural antioxidants from vegetable
and fruit wastes. Food Chem 97:13750.
Soong YY, Barlow PJ 2004. Antioxidant activity and phenolic content of selected fruit seeds.
Food Chem 88:4117.
Sultana B, Anwar F, Przybylski R. 2007. Antioxidant activity of phenolic components present in
barks of Azadirachta indica, Terminalia arjuna, Acacia nilotica, and Eugenia jambolana Lam. trees.
Food Chem 104:110614.
Torunn S, Siv F, Remberg KAL. 2007. Antioxidant capacity and phenolic content of cocoa
beans. Food Chem 100:152330.
Yen GC, Duhand PD, Chaung DY. 2000. Antioxidant of anthraquinones and anthrone. Food
Chem 70:30715.
C852 Journal of Food Science
r
Vol. 77, Nr. 8, 2012