J. Agric. Food Chem.

2003, 51, 667−672 667

Changes in Phenolic Composition and Antioxidant Activity of

Virgin Olive Oil during Frying
SERGIO GOÄ MEZ-ALONSO,† GIUSEPPE FREGAPANE,†
M. DESAMPARADOS SALVADOR,† AND MICHAEL H. GORDON*,‡
Departamento de Quı́mica Analı́tica y Tecnologı́a de los Alimentos, Facultad de Quı́micas,
Universidad de Castilla-La Mancha, Avenida Camilo José Cela, 10, 13071 Ciudad Real, Spain,
and Hugh Sinclair Unit of Human Nutrition, School of Food Biosciences, The University of Reading,
Whiteknights, P.O. Box 226, Reading RG6 6AP, U.K.

The concentration of hydroxytyrosol (3,4-DHPEA) and its secoiridoid derivatives (3,4-DHPEA-EDA

and 3,4-DHPEA-EA) in virgin olive oil decreased rapidly when the oil was repeatedly used for preparing
french fries in deep-fat frying operations. At the end of the first frying process (10 min at 180 °C), the
concentration of the dihydroxyphenol components was reduced to 50-60% of the original value,
and after six frying operations only about 10% of the initial components remained. However, tyrosol
(p-HPEA) and its derivatives (p-HPEA-EDA and p-HPEA-EA) in the oil were much more stable during
12 frying operations. The reduction in their original concentration was much smaller than that for
hydroxytyrosol and its derivatives and showed a roughly linear relationship with the number of frying
operations. The antioxidant activity of the phenolic extract measured using the DPPH test rapidly
diminished during the first six frying processes, from a total antioxidant activity higher than 740 µmol
of Trolox/kg down to less than 250 µmol/kg. On the other hand, the concentration of polar compounds,
oxidized triacylglycerol monomers (oxTGs), dimeric TGs, and polymerized TGs rapidly increased
from the sixth frying operation onward, when the antioxidant activity of the phenolic extract was very
low, and as a consequence the oil was much more susceptible to oxidation. The loss of antioxidant
activity in the phenolic fraction due to deep-fat frying was confirmed by the storage oil and oil-in-
water emulsions containing added extracts from olive oil used for 12 frying operations.

KEYWORDS: Phenols; antioxidant activity; DPPH; virgin olive oil; deep-fat frying

INTRODUCTION elenolic acid (3,4-DHPEA-EDA, 3,4-DHPEA-EA, HPEA-EDA,

Virgin olive oil is the most commonly used cooking fat in and HPEA-EA), which are described as secoiridoids (3-7).
Mediterranean countries, and frying is an important cooking Moreover, significant amounts of the lignans pinoresinol and
technique employed in domestic and industrial food preparation. 1-acetoxypinoresinol are also present (6-9). Some of these
During this operation, due to the high temperature and the compounds possess antioxidant activity (10, 11) and therefore
absorption of oxygen and water, triacylglycerols in the oil suffer improve the oxidative stability of virgin olive oil and extend
a series of reactions, namely hydrolysis, oxidation, isomerization, its shelf life (11-13). Previous studies have used the Folin-
and polymerization (1). Changes in minor components, e.g., Ciocalteu method to investigate the residual total polyphenol
sterols or steryl esters, including the elimination of water or content after frying (14, 15), but this classic method is less
organic acid, also occur. specific and informative than the quantification of individual
Recent studies suggest that the phenolic compounds naturally phenols by HPLC.
contained in virgin olive oil improve its resistance to oxidative The higher oxidative stability of virgin olive oil, compared
deterioration (2). Phenolic antioxidants interrupt the initiation to that of other vegetable oils, is due to both the high oleic acid
and propagation stages of the oxidative chain reaction since they (monounsaturated) and low polyunsaturated fatty acid content
react with lipid radicals to form more stable products (1). The of the triacylglycerols, and also to the level of natural phenolic
main components of the phenolic fraction of virgin olive oil components with antioxidant activity. Within Spanish virgin
are hydroxytyrosol (3,4-DHPEA), tyrosol (HPEA), and their olive oil varieties, Cornicabra and Picual are those with the
derivatives linked to the aldehydic and dialdehydic forms of highest content of both oleic acid and phenolic compounds (16,
17).
* Corresponding author (E-mail [email protected]; telephone The purpose of this work was to study the changes in the
+44 118 3786723; fax + 44 118 9310080).
† Universidad de Castilla-La Mancha. phenols of virgin olive oil and their antioxidant activity
‡ The University of Reading. following several deep-fat frying operations. The antioxidant
10.1021/jf025932w CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/19/2002
668 J. Agric. Food Chem., Vol. 51, No. 3, 2003 Gómez-Alonso et al.

activity of the phenolic fraction of virgin olive oil was quantified at 280 nm using syringic acid as internal standard and the
characterized by its activity in stabilizing oils and emulsions response factors determined by Mateos et al. (6).
against oxidative deterioration, and by the DPPH test. Both oils Analysis of Polar Compounds [adapted from Márquez-Ruiz et al.
and emulsions were used to determine whether possible changes (22)]. Oxidized polar compounds were determined by high-performance
in the polarity of the antioxidants during frying would affect size-exclusion chromatography (HPSEC) with an Agilent 1100 HPLC
(Agilent Technologies, Boeblingen, Germany) equipped with a refrac-
the activity differently in the two media. The DPPH test has
tive index detector (working at 35 °C), using two PL gel columns (300
previously been successfully employed in assessing the anti- × 7.5 mm, 5 µm particle size and 100 and 500 Å pore size, respectively;
oxidant activity of the phenolic extracts from virgin olive oil Agilent, Bracknell, UK) at 30 °C. Tetrahydrofuran at 1 mL/min was
(2, 18, 19). Frying oil deterioration was monitored by the used as the mobile phase. The injection volume was 20 µL, and
formation of oxidized triacylglycerols (monomers) (oxTGs) and monoolein was added as an internal standard.
dimeric and polymerized TGs during frying and under acceler- DPPH Radical Scavenging Effect. DPPH was used as a stable
ated oxidation conditions at 60 °C. radical (23). Several concentrations of SPE phenolic extracts dissolved
in methanol (0.1 mL) were added to a DPPH methanolic solution (2.9
mL, 6 × 10-5 M). The decrease in absorbance of the resulting solution
MATERIALS AND METHODS was then measured at 515 nm at 0, 5, 10, 15 min, and then every 15
Olive Oil Samples and Reagents. Extra virgin olive oil (EVOO) min until 1 h, with a Perkin-Elmer Lambda-Bio spectrometer (Bea-
of the Cornicabra variety was purchased from a Spanish industrial oil consfield, UK). The absorbance was plotted against time, and the
mill (Toledo, Spain). Refined olive oil (OO) was purchased from a percentage of absorbance reduction at 15 min was used as a measure
local retailer (Reading, England). Syringic acid, p-hydroxyphenylacetic of the antioxidant activity of the extract. A calibration curve was
acid, Folin-Ciocalteu reagent, caffeic acid, R-tocopherol (95%), determined using Trolox as an external standard with a range of
monoolein, 2,2-diphenyl-1-picrylhydrazyl radical (DPPH), Trolox, concentrations from 0.19 to 0.93 mM. The equation was remaining
hexanal, and Tween 20 were purchased from Sigma Chemical Co. DPPH (%) ) -132.08[Trolox] + 102.17 (r 2 ) 0.99). The results were
(Poole, UK). All solvents used throughout the study were HPLC or expressed as micromoles of Trolox per kilogram of EVOO.
analytical grade, obtained from Merck (Poole, UK) or Rathburn Emulsion Preparation. Oil-in-water emulsions (30:70, 50 g) were
(Walkerburn, UK). prepared by dissolving Tween 20 (1%) in acetate buffer (0.1 M, pH
5.4) which was cooled in an ice bath. The oil sample was added
Purified Antioxidant-Free Olive Oil (POO) Preparation. Phenols
dropwise while the sample was sonicated with a Vibracell high-intensity
and tocopherols were removed from OO by column chromatography
ultrasonic processor (Sonics & Materials Inc., Danbury, CT) (25). The
using alumina (Merck), according to the method of Yoshida (20).
separation of the oil from emulsions for analysis was performed by
Deep Frying. Potato slices (200 g) were cut into pieces (ap- freezing at -70 °C, thawing, and centrifugation.
proximately 40-50 × 10 × 10 mm) and fried in a DeLonghi domestic Accelerated Oxidation Experiments. Antioxidants and extracts
fryer (model F620), filled initially with EVOO (2 L). The oil was heated were added to purified OO (20 g) before oil storage or to purified OO
to 180 °C, and the potato slices were added and fried for about 10 (15 g) before emulsion preparation. A control sample and samples
min. After each frying operation, the oil was allowed to cool to less containing 0.31 and/or 0.62 mmol‚kg-1 of R-tocopherol or phenolic
than 50 °C. Twelve batches of potatoes were fried with the same oil extract from fresh or from EVOO used for 12 frying operations were
during a 6-day period (two operations per day). Thus, the total frying studied. The mean molecular mass of the phenolic compounds was
period by the end of the experiment was 2 h, with 12 cooling stages in assumed to be 300 for calculation of the extract concentration. Purified
total. Samples of oil (30 g) were removed after each frying operation (antioxidant-free) bulk OO (20 g) and OO-in-water emulsions (50 g)
and stored at -20 °C until analysis. were oxidized at 60 °C in darkness in a 50-mL beaker covered with
Extraction of Phenolic Compounds. A sample of virgin olive oil aluminum foil for the bulk oil or in a 100-mL capped glass bottle for
(2.5 g) was weighed, and when the extract was intended for HPLC the emulsions. The progress of oxidation was monitored by measuring
analysis, 250 µL of a solution of the internal standard (15 mg‚kg-1 of the peroxide value (PV) (AOCS Official Method Cd 8-53) and hexanal
syringic acid in methanol) was added and the solvent was evaporated content by solid-phase microextraction (SPME) and GC as described
with a rotary evaporator at 35 °C under vacuum. The oil was then previously (24). Each experiment was carried out in triplicate.
dissolved in 6 mL of hexane. A diol-bonded phase cartridge (Supelco
Co., Bellefonte, PA) was used to extract the phenolic fraction. The
RESULTS AND DISCUSSION
cartridge was conditioned by eluting with methanol (6 mL) and hexane
(6 mL), and the oil solution was then applied to the SPE column. The Changes in Phenols Composition. The natural phenolic
column was washed with hexane (2 × 3 mL) and with hexane/ethyl compounds in virgin olive oil suffered important changes during
acetate (85:15, v/v; 4 mL), which were run through the cartridge and frying, in agreement with previous literature (14, 15). The initial
discarded. Finally, the phenols were eluted with methanol (15 mL),
concentration of these compounds and their levels after 6 and
and the solvent was removed with a rotary evaporator at 35 °C under
vacuum until dryness. The phenolic residue was dissolved in methanol/ 12 frying operations, determined by SPE, are reported in Table
water (1:1 v/v; 250 µL) for HPLC analysis or in methanol for the DPPH 1. All components decreased in concentration with an increase
test. in the number of frying operations, although the rate of loss
For the accelerated oxidation assay, where a higher amount of depended on chemical structure and antioxidant activity.
phenolic extract was required, phenolic compounds from fresh EVOO The concentration of 3,4-DHPEA and its derivatives 3,4-
or EVOO after frying were extracted by liquid-liquid extraction with DHPEA-EDA and 3,4-DHPEA-EA in virgin olive oil rapidly
aqueous methanol (80:20), as described by Montedoro et al. (21). diminished with the number of frying operations, as shown in
RP-HPLC Determination of Phenols [modified from Mateos et Figure 1. By the end of the first process (10 min at 180 °C),
al., (6)]. HPLC analysis was performed using an HP series 1050 system these components had decreased by 40-50% of their original
equipped with an automatic injector and a diode array detector. A concentration, and after six frying operations less than 10% of
reverse-phase C18 column (250 × 4.6 mm i.d., 5 µm particle size) the original content of these components remained (Table 1).
(Kromasil, Hichrom Ltd., Reading, UK) was used, with an injection
The level of these components was very low after 12 frying
volume of 20 µL and a flow rate of 1.0 mL/min. The mobile phase
was a mixture of water/acetic acid (95:5 v/v) (solvent A), methanol
operations.
(B), and acetonitrile (C). The mobile phase gradient was from 95% The observed trend was consistent with the high antioxidant
(A)-2.5% (B)-2.5% (C) to 34% (A)-33% (B)-33% (C) in 50 min, activity of hydroxytyrosol and its secoiridoid derivatives in
followed by 100% (B) for 15 min to clean the column. Chromatograms virgin olive oil (11, 26), since antioxidants act by reacting
were analyzed at 240, 280, and 335 nm, and phenolic compounds were rapidly with lipid radicals and are thereby consumed. The level
Effects of Frying on Virgin Olive Oil J. Agric. Food Chem., Vol. 51, No. 3, 2003 669

Table 1. Changes in Phenolic Composition of Virgin Olive Oil during

Fryinga
no. of frying operations
compound 0 6 12
hydroxytyrosol 4.88 ± 0.45 0.45 ± 0.01 nd
(3,4-DHPEA)
3,4-DHPEA-EDA 52.49 ± 4.24 3.80 ± 0.40 0.79 ± 0.05
3,4-DHPEA-EA 95.51 ± 7.57 8.81 ± 0.58 5.68 ± 0.64
tyrosol (p-HPEA) 6.59 ± 0.15 5.68 ± 0.02 5.26 ± 0.14
p-HPEA-EDA 96.70 ± 8.52 103.0 ± 0.75 77.46 ± 2.44
p-HPEA-EA 53.09 ± 2.11 29.92 ± 1.38 12.94 ± 0.26
pinoresinol 2.40 ± 0.28 0.80 ± 0.03 0.57 ± 0.00
1-acetoxypinoresinol 8.50 ± 1.34 0.99 ± 0.17 0.73 ± 0.06
elenolic acid 12959 ± 70 1060 ± 45 292 ± 9

aConcentrations expressed as milligrams per kilogram, except for elenolic acid

which is in arbitrary peak area units. nd, not detected.

Figure 2. Changes in the content of tyrosol and its derivatives during

frying. 2, p-HPEA; 9, p-HPEA-EDA; b, p-HPEA-EA; ], sum of the
previous three compounds.

Figure 1. Changes in the content of hydroxytyrosol and its derivatives

during frying. 2, 3,4-DHPEA; 9, 3,4-DHPEA-EDA; b, 3,4-DHPEA-EA;
], sum of the previous three compounds.
of hydroxytyrosol and its derivatives in virgin olive oil was
reported to correlate well with the oxidative stability of the oil, Figure 3. Changes in other polar compounds from the phenolic fraction
as determined by the Rancimat method (7, 12, 13). Moreover, during frying. 9, Elenolic acid; b, pinoresinol; 2, 1-acetoxypinoresinol.
the similar rate of loss of these three compounds agrees with
the results reported by Baldioli et al. (11), who found similar after 12 frying operations decreased by only about 20% of the
antioxidant activity for each of these compounds. The recent initial value, as compared to >95% for hydroxytyrosol and its
report that hydroxytyrosol and its derivatives were lost more secoiridoid derivatives. The loss of p-HPEA-EA was higher,
rapidly during heating at 180 °C from an oil richer in linoleic although its concentration after 12 frying operations was still
acid (Arbequina variety) than from an oil richer in oleic acid nearly 25% of the initial content. These results are consistent
(Picual variety) (27) supports the role of oxidative deterioration with the much higher antioxidant activity of hydroxytyrosol and
in the loss of the antioxidants. However, a contribution of non- its derivatives, as compared to that of the tyrosol family, in
oxidative thermal degradation to antioxidant loss cannot be virgin olive oil, as reported by other authors (11, 26). According
discounted, and some of the phenols may have been lost by to a recent report (27), most of the tyrosol derivatives were also
oxidative coupling, although losses by this route should be small lost by oxidation during prolonged heating at 180 °C, although
due to the low concentration of phenols in the oil. the total heating period was much longer (up to 24 h compared
Another important family of compounds within the phenolic with 2 h in the present frying experiments), and the change in
fraction of virgin olive oil is tyrosol (p-HPEA) and its p-HPEA-EDA was not clear due to coelution of an oxidized
secoiridoid derivatives (p-HPEA-EDA and p-HPEA-EA). This compound with this component.
family showed a completely different behavior during the 12 Elenolic acid is not a phenol, but undoubtedly it is contained
frying processes studied. The reduction in the concentration of in the polar extract of virgin olive oil. The presence of this
this group of compounds was much smaller than that observed compound in a free form is due to the hydrolysis of oleuropein,
for the hydroxytyrosol family and moreover showed an almost ligstroside, and related compounds (6, 21, 28). The concentration
linear relationship with the number of frying operations (Figure of this acid rapidly diminished during frying, as shown in Figure
2). The observed trend is similar to that reported by Brenes et 3, and the acid was present at only about 2% of its initial content
al. (28) in virgin olive oil of the Picual variety during storage after 12 frying operations. This does not mean that this
at room temperature. The p-HPEA and p-HPEA-EDA content compound possessed any antioxidant activity, but the observed
670 J. Agric. Food Chem., Vol. 51, No. 3, 2003 Gómez-Alonso et al.

Figure 4. Changes in the antioxidant activity of the phenolic extract,

assessed by the DPPH assay, and in the levels of oxidized polar
compounds formed during frying. 0, TAA; b, OxTGs; [, PTGs + DTGs. Figure 5. Oxidation at 60 °C of bulk oil containing phenolic extracts from
fresh and used virgin olive oil. 9, Control; b, R-tocopherol (0.62 mM);
decrease was probably due to rapid oxidation. Briante et al. (29) 2, fresh oil extract (0.62 mM); 4, extract from fresh virgin olive oil (0.31
showed that elenolic acid was a very poor antioxidant compared mM); [, extract from virgin olive oil used for 12 frying operations (0.62
to hydroxytyrosol. mM); ], extract from virgin olive oil used for 12 frying operations (0.31
The lignans pinoresinol and 1-acetoxypinoresinol were de- mM).
scribed as major components of the phenolic fraction of virgin
olive oil (8). According to the literature, these compounds frying operations. The loss of this family of antioxidants would
possess in vitro antioxidant activity (30), and they may be able allow the increase in the rate of formation of oxidized polar
to inhibit lipid peroxidation in vivo if they are absorbed (31). compounds in the oil.
As was also observed for elenolic acid, the content of these The nutritional quality of the fried potatoes is likely to
compounds fell drastically in the first frying operation, with a deteriorate after six frying operations as the level of oxidized
much lower reduction rate in subsequent frying operations polar compounds increases because oil absorbed by fried
(Figure 3). The increased stability of some lignans compared potatoes from poor-quality oil contains high levels of oxidized
to that of hydroxytyrosol derivatives during the frying of polar compounds (32), which showed negative physiological
potatoes is consistent with the recent report covering changes and biochemical effects in rats fed high doses of this polar
during dry heating at 180 °C (27). material (33, 34). Nevertheless, the levels of TPM (9%) and
Changes in the Antioxidant Activity. The antioxidant DTGs and PTGs (4.4%) in the olive oil after 12 frying
activity of the phenolic extract, determined by the DPPH test, operations (Figure 4) were much lower than the upper limits
rapidly decreased progressively during the first six frying of 24% and 12% for TPM and DTGs + PTGs, respectively,
processes, from a total antioxidant activity (TAA) higher than recommended by the German Society for Fat Science (35).
740 µmol of Trolox/kg down to less than 250 µmol/kg, as shown To verify whether the residual phenolic compounds contained
in Figure 4. During subsequent frying operations, the total in the olive oil at the end of the 12 frying operations still
antioxidant activity of the phenolic extract continued to decrease maintained any antioxidant effects, the tocopherol-free phenolic
progressively, but the rate of reduction was much lower. To extract from this oil was added to a purified antioxidant-free
confirm that the method of isolation with the SPE cartridge was olive oil (POO). This olive oil was stored in the dark at 60 °C,
able to extract the antioxidant components completely for both and the peroxide value was periodically measured during the
fresh and used virgin olive oil, liquid-liquid extraction of the course of the oxidation process. The stability of this oil was
phenolic fraction was performed with the original virgin olive compared with that of samples of the oil without additives
oil and the oil after 12 frying batches, and the DPPH scavenging (control) and with the addition of R-tocopherol (0.62 mM) and
activity was nearly the same as for the samples isolated by SPE. two levels of phenolic extract obtained from the fresh virgin
The TAA for fresh oil was 693 and 746 µmol of Trolox/kg for olive oil (0.62 and 0.31 mM). These concentrations were chosen
liquid-liquid and solid-phase extraction, respectively, and 139 to reproduce the original concentration of phenolic extract and
and 144 µmol/kg respectively for oil after 12 frying operations. half the original concentration in the Cornicabra olive oil.
Since large amounts of phenolic extract were required for The stability of the olive oil samples containing the additives
accelerated oxidation experiments, liquid-liquid extraction was is shown in Figure 5. It is clear that the addition of fresh virgin
used to isolate the extract for this application, whereas SPE was olive oil phenolic extract gave a highly stable oil, and that after
used for the isolation of the extract for HPLC analysis due to more than 300 h of storage at 60 °C, the peroxide value of the
its speed and convenience. oil remained lower than 10 mequiv/kg. The oil containing
The concentration of total polar material (TPM), including phenolic extract from fresh virgin olive oil was even more stable
oxidized triacylglycerol monomers (oxTGs), dimeric TGs than that obtained by the addition of an equimolar concentration
(DTGs), and polymerized TGs (PTGs) (Figure 4), increased of R-tocopherol when assessed by PV, but the sample containing
rapidly from the sixth frying operation onward, which is R-tocopherol contained less hexanal than the sample containing
consistent with the low antioxidant activity of the phenolic phenolic extract from fresh virgin olive oil after 328 h of storage
extract after this point, with the consequence that the oil was (Table 2). However, the phenolic extract from the oil used for
more susceptible to oxidation. These findings are consistent with 12 frying operations did not contribute any increased stability
the low residual content of the main antioxidant compounds, to the oil. Indeed, a small pro-oxidant effect was observed when
hydroxytyrosol and its derivatives, in virgin olive oil after six changes in PV for this sample were compared to those of the
Effects of Frying on Virgin Olive Oil J. Agric. Food Chem., Vol. 51, No. 3, 2003 671

Table 2. Hexanal Content after Storage at 60 °C 2), showed a large increase in this oxidation product in stored
samples containing the phenolic extract after 12 frying opera-
hexanal contenta tions as compared with the POO with added fresh phenolic
prepared olive oil bulk oilb emulsionc extract (50 times higher) for both bulk oils and emulsions. This
fresh POO 164 ± 14 confirmed the absence of antioxidant activity in the extract after
control 26 260 ± 3104 16 119 ± 1803 12 frying operations, since the hexanal content was similar to
R-tocopherol (0.62 mM) 286 ± 52 351 ± 53 that in the control sample.
fresh Cornicabra virgin olive 432 ± 96 3434 ± 666
oil extract (0.62 mM) Virgin olive oil shows a good behavior during frying in terms
12 frying extract (0.62 mM) 21 317 ± 2565 16 066 ± 1533 of resistance to formation of polar compounds and TPMs. This
was found in the current frying experiment and is similar to
a Expressed as arbitrary peak area units. b Bulk oil after 328 h at 60 °C.
the results found by other authors under similar conditions (15,
c Emulsion after 208 h at 60 °C. 36). The low linoleic acid content of the oil is significant in
contributing to this effect, and the phenolic components may
contribute to the oil stability in the early stages of frying.

LITERATURE CITED

(1) Frankel, N. Lipid Oxidation; The Oily Press Ltd.: Dundee,

Scotland, 1998.
(2) Gordon, M. H.; Paiva-Martins, F.; Almeida, M. Antioxidant
activity of hydroxytyrosol acetate compared with that of other
olive oil polyphenols. J. Agric. Food Chem. 2001, 49, 2480-
2485.
(3) Vázquez, A.; Janer, C.; Janer, M. L. Determinación de los
polifenoles del aceite de oliva. Grasas Aceites 1973, 24, 350-
357.
(4) Montedoro, G. F.; Servilli, M.; Baldioli, M.; Selvaggini, R.;
Miniati, E.; Macchioni, A. Simple and hydrolyzable phenolic
compounds in olive oil. Note 3. Spectroscopic characterization
Figure 6. Oxidation at 60 °C of oil in water emulsions containing phenolic of the secoiridoid derivates. J. Agric. Food Chem. 1993, 41,
2228-2234.
extracts from fresh and used virgin olive oil. 9, Control; b, R-tocopherol
(5) Brenes, M.; Garcı́a, A.; Garcı́a, P.; Rios, J. J.; Garrido, A.
(0.62 mM); 2, fresh oil extract (0.62 mM); 4, fresh oil extract (0.31 mM);
Phenolic compounds in Spanish olive oils. J. Agric. Food Chem.
[, extract from virgin olive oil used for 12 frying operations (0.62 mM);
1999, 47, 3535-3540.
], extract from virgin olive oil used for 12 frying operations (0.31 mM). (6) Mateos, R.; Espartero, J. L.; Trujillo, M.; Rios, J. J.; León-
Camacho, M.; Alcudia, F.; Cert, A. Determination of phenols,
control oil, and the pro-oxidant effect was enhanced when the flavones and lignans in virgin olive oils by solid-phase extraction
higher level of extract was present. The pro-oxidant effect was and high-performance liquid chromatography with diode array
not evident from the hexanal determinations, but the lack of an ultraviolet detection. J. Agric. Food Chem. 2001, 49, 2185-
antioxidant effect was confirmed. The pro-oxidant effect was 2192.
probably caused by the presence of other compounds in the (7) Gómez-Alonso, S.; Salvador, M. D.; Fregapane, G. Phenolic
extract, like the oxidized polar compounds that favor the Compounds Profile of Cornicabra Virgin Olive Oil. J. Agric.
oxidation process. In fact, the level of oxidized triacylglycerols, Food Chem. 2002, 50, 6812-6817.
and triacylglycerol dimers and polymers in the phenolic extract (8) Owen, R. W.; Mier, W.; Giacosa, A.; Hull, W. E.; Spiegelhalder,
from oil used for 12 frying operations was increased by 11 and B.; Bartsch, H. Identification of lignans as major components
20 times, respectively, compared to their content in the fresh in the phenolic fraction of olive oil. Clin. Chem. 2000, 46, 976-
virgin olive oil (Figure 4). 988.
(9) Brenes, M.; Hidalgo, F. J.; Garcı́a, A.; Rios, J. J.; Garcı́a, P.;
The fact that the phenolic extract from the frying oil used Zamora, R.; Garrido, A. Pinoresinol, and 1-acetoxypinoresinol,
for 12 frying operations did not show any antioxidant activity two new phenolic compounds identified in olive oil. J. Am. Oil
in stored oil samples indicates that residual antioxidant activity Chem. Soc. 2000, 77, 715-720.
detected by the DPPH method was due to compounds that react (10) Gutfinger, T. Polyphenols in olive oils. J. Am. Oil Chem. Soc.
with DPPH radicals but do not interrupt the lipid peroxidation 1981, 58, 966-968.
chain reaction. Tyrosol and its derivatives, which are known (11) Baldioli, M.; Servilli, M.; Perretti, G.; Montedoro, G. F.
not to possess antioxidant activity in stored oil samples, may Antioxidant activity of tocopherols and phenolic compounds of
have been responsible for this misleading residual activity in virgin olive oil. J. Am. Oil Chem. Soc. 1996, 73, 1589-1593.
the DPPH test. (12) Montedoro, G. F.; Servilli, M.; Baldioli, M.; Miniati, E. Simple
The antioxidant behavior of the extracts in an oil-in-water and hydrolyzable phenolic compounds in olive oil. Note 2. Initial
emulsion was similar to that observed for the bulk oil. The main characterization of the hydrolyzable fraction. J. Agric. Food
Chem. 1992, 40, 1577-1580.
difference was that, in this case, the phenolic extract from oil
(13) Caponio, F.; Alloggio, V.; Gomes, T. Phenolic compounds of
used for 12 frying operations did not show any pro-oxidant effect
virgin olive oil: influence of paste preparation techniques. Food
when evaluated by PV measurements (Figure 6), possibly due Chem. 1999, 64, 203-209.
to the oxidized compounds partitioning into the aqueous phase (14) Pellegrini, N.; Visioli, F.; Buratti, S.; Brighenti, F. Direct analysis
of the emulsion. of total antioxidant activitty of olive oil and studies on the
The analysis of hexanal in the volatile fraction of oil or influence of heating. J. Agric. Food Chem. 2001, 49, 2532-
emulsion samples stored for 328 and 208 h, respectively (Table 2538.
672 J. Agric. Food Chem., Vol. 51, No. 3, 2003 Gómez-Alonso et al.

(15) Andrikopoulos, N. K.; Kalogeropoulos, N.; Falirea, A.; Bar- (27) Brenes, M.; Garcı́a, A.; Dobarganes, M. C.; Velasco, J.; Romero,
bagianni, M. N. Performance of virgin olive oil and vegetable C. Influence of thermal treatments simulating cooking processes
shortening during domestic deep-frying and pan-frying of on the polyphenol content in virgin olive oil. J. Agric. Food.
potatoes. Int. J. Food Sci. Technol. 2002, 37, 177-190. Chem. 2002, 50, 5962-5967.
(16) Aparicio, R.; Roda, L.; Albi, M. A.; Gutiérrez, F. Effects of (28) Brenes, M.; Garcı́a, A.; Garcı́a, P.; Garrido, A. Acid hydrolysis
various compounds on virgin olive oil stability measured by of secoiridoid aglycons during storage of virgin olive oil. J. Agric.
Rancimat. J. Agric. Food Chem. 1999, 47, 4150-4155. Food. Chem. 2001, 49, 5609-5614.
(17) Salvador, M. D.; Aranda, F.; Gómez-Alonso, S.; Fregapane, G. (29) Briante, R.; La Cara, R.; Tonziello, M. P.; Febbraio, F.; Nicci,
Cornicabra virgin olive oil: a study of five crop seasons. R. Antioxidant activity of the main bioactive derivatives from
Composition, quality and oxidative stability. Food Chem. 2001,
Oleuropein hydrolysis by hyperthermophilic β-glycosidase. J.
74, 267-274.
Agric. Food Chem. 2001, 49, 3198-3203.
(18) Mosca, L.; De Marco, C.; Visioli, F.; Cannella, C. Enzymatic
(30) Owen, R. W.; Mier, W.; Giacosa, A.; Hull, W. E.; Spiegelhalder,
assay for the determination of olive oil polyphenol content: assay
conditions and validation of the method. J. Agric. Food Chem. B.; Bartsch, H. Phenolic compounds and squalene in olive oils:
2000, 48, 297-301. the concentration and antioxidant potential of total phenols,
(19) Frankel, E. N.; Meyer, A. S. Review. The problems of using simple phenols, secoiridoids, lignans and squalene. Food Chem.
one-dimensional methods to evaluate multifunctional food and Toxicol. 2000, 38, 647-659.
biological antioxidants. J. Sci. Food Agric. 2000, 80, 1925- (31) Kang M. H.; Mait, M.; Tsujihara, N.; Osawa, T. Sesamolin
1941. inhibits lipid peroxidation in rat liver and kidney. J. Nutr. 1998,
(20) Yoshida, H. Influence of fatty acids of different unsaturation in 128, 1018-1022.
the oxidation of purified vegetable oils during microwave (32) Gordon, M. H.; Kourimská, L. The effects of antioxidants on
irradiation. J. Sci. Food Agric. 1993, 62, 41-47. changes in oils during heating and deep frying. J. Sci. Food
(21) Montedoro, G. F.; Servilli, M.; Baldioli, M.; Miniati, E. Simple Agric. 1995, 68, 347-353.
and hydrolyzable phenolic compounds in olive oil. Note 1. Their (33) Billek, G. Die Veränderungen von Nahrungsfetten bei höheren
extraction, separation, and quantitative and semiquantitative Temperaturen. Fat Sci. Technol. 1992, 94, 161-172.
separation and evaluation by HPLC. J. Agric. Food Chem. 1992, (34) Lamboni, C.; Perkins, E. G. Effects of dietary heated fats on rat
40, 1571-1576. level enzyme activity. Lipids 1996, 31, 955-962.
(22) Márquez-Ruiz, G.; Jorge, N.; Martı́n-Polvillo, M.; Dobarganes, (35) German Society for Fat (DGF): Recommendations of the 3rd
M. C. Rapid, quantitative determination of polar compounds in International Symposium on deep-fat frying-optimal operation,
fats and oils by solid-phase extraction and size-exclusion 20-21 March 2000, Hagen (Germany). Eur. J. Lipid Sci.
chromatography using monostearin as internal standard. J. Technol. 2000, 102, 594.
Chromatogr. A 1996, 749, 55-60.
(36) Andrikopoulos, N. K.; Dedoussis, G. V. Z.; Tzamtzis, V.; Chiou,
(23) Brand-Williams, W.; Cuvelier, M. E.; Berset, C. Use of a free
A.; Boskou, G. Evaluation of medium polarity materials isolated
radical method to evaluate antioxidant activity. Lebensm.-Wiss.
from fried edible oils by RP-HPLC. Eur. J. Lipid Sci. Technol.
Technol. 1995, 28, 25-30.
(24) Keceli, T.; Gordon, M. H. The antioxidant activity and stability 2002, 104, 110-115.
of the phenolic fraction of green olives and extra virgin olive
oil. J. Sci. Food Agric. 2001, 81, 1391-1396. Received for review September 3, 2002. Revised manuscript received
(25) Roedig-Penman, A.; Gordon, M. H. Antioxidant properties of November 14, 2002. Accepted November 18, 2002. This research was
myricetin and quercetin in oil and emulsions. J. Am. Oil Chem. partially supported by the CICYT (Comisión Interministerial de Ciencia
Soc. 1998, 75, 169-180. y Tecnologı́a; AGL 2001-0906).
(26) Papadopoulos, G. K.; Boskou, D. Antioxidant effect of natural
phenols on olive oil. J. Am. Oil Chem. Soc. 1991, 68, 669-671. JF025932W