LWT - Food Science and Technology 58 (2014) 470e477

Contents lists available at ScienceDirect

LWT - Food Science and Technology

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

Release and antioxidant activity of carvacrol and thymol from

polypropylene active packaging films
Marina Ramos a, *, Ana Beltrán a, Mercedes Peltzer a, Artur J.M. Valente b,
María del Carmen Garrigós a
a
Analytical Chemistry, Nutrition & Food Sciences Department, University of Alicante, 03080, Alicante, Spain
b
Department of Chemistry, University of Coimbra, 3004-535, Coimbra, Portugal

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

Article history: The migration of antioxidant (AO) agents, carvacrol and thymol, from polypropylene (PP) packaging films
Received 10 October 2013 containing the studied compounds at 80 g/kg separately and an equimolar mixture of them into food
Received in revised form simulants was investigated. Fast and reliable analytical procedures were developed and validated for the
20 March 2014
analysis of the studied AOs in food simulants. For aqueous food simulants, solid phase extraction fol-
Accepted 10 April 2014
Available online 25 April 2014
lowed by GCeMS analysis was performed. Fatty food simulants were directly analysed by GCeMS and
HPLC-UV for isooctane and ethanol 950 mL/L, respectively. The release of AOs from the films was
dependent on the type of food stimulant and AO incorporated. In particular, high levels of migration were
Keywords:
Active packaging
obtained for both AOs into isooctane, showing thymol higher migration. The release kinetics of AOs from
Carvacrol PP films showed a Fichian behaviour with diffusion coefficients ranging from 1 to 2
1014 m2/s; except
Thymol for the diffusion into isooctane where 4-6 higher values were obtained. The antioxidant activity of
Migration migration extracts was confirmed by the DPPH method, showing thymol a higher antioxidant capacity
Diffusion especially into isooctane with a 42.2% of inhibition. The obtained results suggest that carvacrol and
thymol show a potential use as AOs for active packaging for extending the shelf-life of food products.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction aromatic plant extracts (Dopico-García et al., 2011; Lopez-de-

Dicastillo et al., 2011) and polyphenols from natural oils (Park
Antioxidant active food packaging is a growing alternative to et al., 2012; Peltzer, Wagner, & Jiménez, 2009). The principal con-
common procedures to protect sensitive oxidation of food such as stituents of oregano essential oil, thymol and carvacrol, exhibit a
the addition of antioxidants directly in food samples in combina- high antioxidant activity as it has been reported (Tomaino et al.,
tion with vacuum or modified atmosphere (Lopez-de-Dicastillo 2005); and they are generally recognized as safe (possess “GRAS”
et al., 2011). These systems are usually based on materials in status) and as flavouring substances according to European Com-
which some additives showing antioxidant properties are added mission Decision 2002/113/EC. Their antioxidant activity can be
directly into the polymer matrix. These additives can play a double evaluated by using diverse methods, such as DPPH (2,20 -diphenyl-
role: a food protection by their controlled release, in particular 1-picrylhydrazyl), which is usually used due to its simple, rapid,
avoiding fat and pigment components oxidation (Del Nobile et al., sensitive, and reproducible procedure (Ozcelik, Lee, & Min, 2003).
2009); and to protect the polymer from degradation during pro- The release rate of AOs from the packaging material can be
cessing. In fact, the addition of antioxidants to polyolefins is a evaluated by using migration studies, which are usually performed
common practice during film manufacturing (Siró et al., 2006; using food simulants and conditions specified in European food
Tovar, Salafranca, Sanchez, & Nerin, 2005). packaging regulations (EC, 2011; Kuorwel, Cran, Sonneveld, Miltz, &
The new trends in using natural additives have produced a clear Bigger, 2013). Migration is the result of diffusion, dissolution and
increase in the number of studies based on natural active com- equilibrium processes involving the mass transfer of low molecular
pounds, such as a-tocopherol (Barbosa-Pereira et al., 2013), mass compounds initially present in the package into a food sample
or food simulant; and it is often described by Fick’s second law
(Manzanarez-López, Soto-Valdez, Auras, & Peralta, 2011). Chro-
* Corresponding author. Tel.: þ34 965903400x3117; fax: þ34 965903697. matographic methods are usually used for identification and
E-mail address: marina.ramos@ua.es (M. Ramos). quantification of migrated compounds (Salafranca, Pezo, & Nerín,

http://dx.doi.org/10.1016/j.lwt.2014.04.019
0023-6438/Ó 2014 Elsevier Ltd. All rights reserved.
 M. Ramos et al. / LWT - Food Science and Technology 58 (2014) 470e477 471

2009). Also, concentration and/or isolation of analytes into a suit- ratio around 6 dm2/L). A blank test for each simulant was also
able solvent may be performed prior to chromatographic analysis carried out.
by the use of sample preparation and purification techniques such
as solid phase extraction (SPE) in order to improve detection and 2.3.2. Migration kinetics
quantification (Burman, Albertsson, & Höglund, 2005; Ridgway, In order to study the release of carvacrol and thymol during a
Lalljie, & Smith, 2007). suitable period of time (15 days), a kinetic study was performed
In a previous study, the effect of the addition of carvacrol and using acetic acid 30 g/L, ethanol 100 mL/L, ethanol 950 mL/L and
thymol at different concentrations to PP films on the thermal, isooctane as food simulants, at the same temperature conditions
structural, mechanical and functional properties was studied as described in Section 2.3.1. Extract samples were taken at 2, 6, 12, 24,
well as the evaluation of the antimicrobial activity against two 48 h and 5, 10 and 15 days in triplicate.
typical food born bacteria, E. coli and S aureus (Ramos, Jiménez, The migration process is described by the kinetic of the diffusion
Peltzer, & Garrigós, 2012). It was concluded that the addition of of the migrant in the film and it is expressed by the diffusion co-
both compounds at 80 g/kg showed some potential to be used as efficient, D (m2/s) (Manzanarez-López et al., 2011). Considering the
active additives in PP formulations; showing thymol higher inhi- case of limited packaging, limited food, where migration occurs
bition against the studied bacteria, leading to higher antimicrobial from a limited volume packaging film into a well-mixed limited
activity. The aim of this study was to evaluate the release of these volume of food, the diffusion coefficients of AOs can be determined
compounds from PP films into different aqueous and fatty food by using a release kinetic model based in the Fick’s second law
simulants; including a kinetics diffusion study and the evaluation of (Equation (1)). In this case, the food sample initially does not
the antioxidant efficiency by the DPPH method. contain any migrant, and as migration occurs, the concentration of
migrant in the food increases from zero CF,0 to its equilibrium value
2. Experimental CF,N. Equation (1) is the most rigorous general model for describing
the migration controlled by Fickian diffusion in a packaging film
2.1. Chemicals (Chung, Papadakis, & Yam, 2002; Crank, 1975):
" #
All reagents used were of analytical or chromatographic grade XN
MF;t 2að1 þ aÞ Dq2n t
and were purchased from Panreac (Barcelona, Spain). Standards of ¼ 1 exp (1)
MF;N n¼1
1 þ a þ a2 q2n L2P
carvacrol (98%), thymol (99.5%), and 2,2-diphenyl-1-
picrylhydrazyl (DPPH, 95%) were acquired from SigmaeAldrich
Inc. (St. Louis, MO). Ultrapure water was obtained from a Millipore where MF,t and MF,N are the total amount of diffusing substance
Milli-Q system (Millipore, Bedford, MA, USA). Polypropylene PP released by the film at time t and after infinite time, respectively; LP
ECOLEN HZ10K pellets (Hellenic Petroleum, Greece) was kindly is the film thickness; qn are the non-zero positive roots of
supplied by Ashland Chemical Hispania (Barcelona, Spain). tanqn¼a,qn; and a is the partition expressed as:
  
a ¼ KFP VF Vp (2)
2.2. Films preparation
where VF and VP are the volumes of the simulant and the polymer,
PP active films were obtained by melt blending at 190  C for
respectively; and KFP is the partition coefficient of the active com-
6 min at 50 rpm followed by compression moulding at 190  C in a
pound between the simulant and the polymer.
hot press, according to a method previously developed (Ramos
A simplified migration model derived from Equation (1) was
et al., 2012). Three active formulations were obtained: PP con-
proposed by Chung et al. useful for linear regression analysis
taining 80 g/kg of thymol (PPT8) or carvacrol (PPC8); and PP with
(Chung et al., 2002):
an equimolar mixture of both additives at 80 g/kg (PPTC8) to study
a possible additive effect of both compounds. An additional sample  0:5
1 MF;t
1 D0:5 0:5 1
without any active compound was also prepared as control (PP0).  $ ¼  $t þ 0:5 (3)
The film thickness was measured using a micrometre (Mitutoyo, p a MP;0 a$LP p
Japan) at five random positions. The average thickness of PP films
incorporated with thymol (PPT8), carvacrol (PPC8) and both addi- where MP,0 is the initial amount of migrant in the packaging film
tives (PPTC8) was found to be 185  3, 192  6 and 190  4 mm, (for a complete migration MP,0 ¼ MF,N). Thus, the diffusion coeffi-
respectively. The final appearance of the films was completely cient can be directly computed from the fitting of Equation (3) to
transparent and homogenous. experimental migration data.

2.3. Migration study 2.4. Analysis of released antioxidants into food simulants

2.3.1. Release tests The amount of released AOs into aqueous food simulants was
The release of AOs from PP films was performed into five food analysed by GCeMS with a previous extraction and concentration
simulants according to European Standard EN 13130-2005 (UNE- step by SPE on an octadecyl cartridge (C18, 500 mg, 6 mL) (Teknok-
EN, 2005): distilled water (A), acetic acid 30 g/L (B), and ethanol roma, Barcelona, Spain). A Büchi V-700 vacuum system (Flawil,
100 mL/L (C) were used as aqueous food simulants; whereas Switzerland) and a vacuum manifold from Teknokroma (Barcelona,
ethanol 950 mL/L and isooctane were employed as fatty food Spain) were used for SPE sample processing. The cartridge was pre-
simulants. viously conditioned with 4 mL methanol and 4 mL distilled water at
Migration studies were conducted in triplicate at 40  C for 10 5 mL/min. Then, the extract was loaded and elution of the AOs was
days in an oven (J.P. Selecta, Barcelona, Spain), except for isooctane carried out with 4 mL dichloromethane (1 mL/min). On the other
studies which were performed at 20  C and 50% relative humidity hand, extracts obtained from isooctane and ethanol 950 mL/L were
for 2 days in a climatic chamber (Dycometal, Barcelona, Spain). directly analysed by GCeMS and HPLC-UV, respectively.
Double-sided, total immersion migration tests were performed Stock (4000 mg/kg) and working solutions of each AO were
with 12 cm2 of films and 20 mL of each simulant (area-to-volume prepared in the appropriate solvent (dichloromethane, isooctane or
472 M. Ramos et al. / LWT - Food Science and Technology 58 (2014) 470e477

ethanol 950 mL/L) depending on the food simulant and chromato- 2.5. Statistical analysis
graphic technique used and stored in a freezer. Carvacrol and thymol
quantification was performed using external calibration in triplicate. One way analysis of variance (ANOVA) was applied on DPPH
experimental data with the aid of the statistical program “Statgraphics
2.4.1. GCeMS Analysis Centurion program v.16.1.18 (StatPoint, Inc., Warrenton, USA)” and
A Perkin Elmer TurboMass Gold GCeMS (Boston, MA, USA) significant differences among sample data were recorded at P < 0.05
operating in electronic impact ionisation mode (70 eV) with a SPB- according to Tukey’s post hoc test (Barbara & Tabachnick, 2013).
5 capillary column (30 m
0.25 mm
0.25 mm; Supelco, Belle-
fonte, PA) was used. The column temperature was programmed 3. Results and discussion
from 60  C (1 min) to 120  C (1 min) at 10  C/min and to 150  C at
2  C/min (2 min). Helium was used as carrier gas at 1 mL/min. Ion 3.1. Validation of the developed methods
source and GCeMS transfer line temperatures were 250 and 270  C,
respectively. Injector temperature was 270  C and 1 mL of extracts The analytical methods developed in this study were validated
were injected (split mode 1:100). by assessing the main analytical characteristics: linearity, precision
Identification of thymol and carvacrol were performed in full (repeatability), detection (LOD) and quantification (LOQ) limits and
scan mode (m/z 30e550) by a combination of NIST mass spectral accuracy (recovery test).
library and retention times of standard compounds. Quantification Linear ranges were calculated with five calibration points, each
of AOs was performed by using selected ion monitoring (SIM) mode one in triplicate (0.15e2.10 mg/kg in dichloromethane for SPE-GC-
focused on m/z 91, 135 and 150. Retention times obtained for MS method and aqueous food simulants; 0.15e4.00 mg/kg for
thymol and carvacrol were 10.7 and 11.0 min, respectively. isooctane and 950 mL/L ethanol (v/v) for direct GCeMS and HPLC-
UV analyses, respectively). The calculated calibration curves gave
an acceptable level of linearity for AOs and studied methods with
2.4.2. HPLC-UV Analysis
determination coefficients (R2) ranging between 0.9963 and
A Shimadzu LC-20A liquid chromatograph equipped with UV
0.9989, as shown in Table 1.
detector and a LiChrospher 100 RP18 column
Repeatability was evaluated by analysing three replicates of
(250 mm
5 mm
5 mm, Agilent Technologies) was used. The
standard solutions processed the same day. All methods showed
mobile phase consisted of acetonitrile: distilled water, 40:60 (v:v)
similar results for relative standard deviation (RSD) values which
at 1 mL/min 20 mL of sample were injected. Detection of carvacrol
were lower than 10%. LOD and LOQ values were determined by
and thymol was performed at 274 nm with retention times of 18.9
using regression parameters from the calibration curves (3 Sy/x/a
and 21.0 min, respectively.
and 10 Sy/x/a, respectively; where Sy/x is the standard deviation of
the residues and a is the slope). As it can be seen in Table 1, lower
2.4.3. Determination of antioxidant activity values for LOD and LOQ were obtained for thymol by the HPLC-UV
The antioxidant activity of AOs released into food simulants was method. On the other hand, carvacrol showed lower values for
analysed in terms of radical scavenging ability, using the stable these parameters considering the SPE-GC-MS method. As a result,
radical DPPH method as proposed by Byun et al. (Byun, Kim, & the LODs and LOQs values obtained for AOs ranged between 0.16
Whiteside, 2010) with some modifications. An aliquot of 100 mL and 0.22 mg/kg and between 0.50 and 0.74 mg/kg, respectively.
of each simulant extract was mixed with 3.9 mL of a methanolic Recovery tests for the SPE-GC-MS method were accomplished,
solution of DPPH (23 mg/L) in a capped cuvette. The mixture was in triplicate, in order to evaluate accuracy, by spiking aqueous food
shaken quickly at room temperature and the absorbance of the simulants with known amounts of each AO at three concentration
solution was measured immediately at 517 nm every 1 min until levels (0.03, 0.27 and 2.60 mg). A working solution containing both
the absorbance value was stabilized (200 min), by using a Biomate- AOs (4000 mg/kg) in methanol was used. Satisfactory results were
3 UVeVIS spectrophotometer (Thermospectronic, USA). All ana- obtained for mean recoveries at all levels tested (Table 2), ranging
lyses were performed in duplicate. from 86.7 to 108.2% with RSD values between 2.1 and 11.0%. In
The ability to scavenge the stable radical DPPH was calculated as conclusion, the results obtained for methods validation were
percent of inhibition (I %) using the following Equation (4): considered acceptable for the determination of carvacrol and
h . i thymol migration in aqueous and fatty food simulants.
Ið%Þ ¼ AControl  ASample AControl $100 (4)
3.2. Release of AOs into food simulants
where AControl and ASample are the absorbances of the control at
t ¼ 0 min (using methanol instead of sample) and of the tested Both AOs were readily released into aqueous and fatty food
sample at t ¼ 200 min, respectively. simulants from all PP films (Table 3). Similar behaviour was

Table 1
Main analytical parameters obtained for the studied AOs using the optimized methods (n ¼ 3).

Analyte Method Parameter

Slope  SD Intercept  SD Linearity (R2)a LOD (mg/kg)b LOQ (mg/kg)c

Carvacrol SPE-GC-MS 18416  1907 3881  2372 0.9968 0.16 0.54

Direct HPLC-UV 13510  333 1195  82 0.9963 0.20 0.66
Direct GCeMS 2241  721 15103  320 0.9982 0.22 0.73
Thymol SPE-GC-MS 19792  1120 3868  2377 0.9972 0.15 0.50
Direct HPLC-UV 13936  177 1091  43 0.9989 0.10 0.34
Direct GCeMS 1852  701 14568  316 0.9981 0.22 0.74
a
Number of calibration points ¼ 5. Linear range: 0.15e2.10 (SPE-GC-MS); 0.15e4.00 (Direct HPLC-UV and GCeMS).
b
Calculated for 3 Sy/x.
c
Calculated for 10 Sy/x.
 M. Ramos et al. / LWT - Food Science and Technology 58 (2014) 470e477 473

Table 2
Mean recoveries (%) and R.S.D. values (%) in parentheses obtained for each AO in
aqueous simulants by SPE-GC-MS (n ¼ 3).

Analyte Simulant Spiking level (mg)

0.03 0.27 2.60

Carvacrol Distilled water 98.1 (5.4) 94.7 (9.7) 108.2 (2.1)

Ethanol 100 mL/L 95.2 (4.3) 100.3 (3.3) 89.8 (2.4)
Acetic acid 30 g/L 99.4 (6.5) 88.0 (3.2) 97.2 (10.9)
Thymol Distilled water 96.8 (5.8) 101.0 (3.6) 106.1 (2.4)
Ethanol 100 mL/L 94.1 (3.8) 99.1 (3.4) 88.4 (2.5)
Acetic acid 30 g/L 94.8 (4.9) 86.7 (3.2) 95.8 (11.0)

observed for thymol and carvacrol migration under the same

migration conditions. However, thymol showed higher migration
tendencies than carvacrol for distilled water. Also, the amount of
active additives released into fatty food simulants was higher than
those obtained for the aqueous ones. In particular, the highest
migration levels were obtained into isooctane at 20  C during 2
days compared with the rest of simulants where 40  C and 10 days
were used. This phenomenon might result from the higher affinity Fig. 1. Radical scavenging activity measured by DPPH, expressed as percent of inhi-
of the non-polar PP to the also non-polar isooctane than to the bition, obtained for migration extracts from the three formulations studied (isooctane:
highly polarity of ethanol 950 mL/L or other polar food simulants 20  C, 2 days; rest of simulants: 40  C, 10 days) (m  SD, n ¼ 3). Different letters
represent significant difference at P < 0.05.
used, therefore showing diffusion behaviour near to extraction
rather than migration, ultimately leading to high migration values.
The higher migration observed into fatty food simulants could diffusing compound (Haider & Karlsson, 2000; Reynier, Dole,
be also attributed to two factors: the higher solubility of migrated Humbel, & Figenbaum, 2001). This behaviour was also observed
AOs into these solvents and the phenomenon of swelling of the for Mastromatteo et al. who demonstrated that the release of
polymer matrix when the films come into contact with simulants thymol from a swelling homogeneous polymeric network could be
(Suppakul, Sonneveld, Bigger, & Miltz, 2011). Tehrany, Mouawad, & viewed as a result of the water diffusion from the outer water so-
Desobry (2007) indicated that migrant polarity can be a predomi- lution into the polymeric matrix, the macromolecular matrix
nant controlling factor and that a simulant with similar high po- relaxation and the diffusion of the active compound from the
larity could have a great effect on sorption. In this sense, swollen polymeric network into the outer water solution
partitioning depends on the polarity and solubility of the migrant (Mastromatteo, Barbuzzi, Conte, & Del Nobile, 2009). Also, it has to
in the food simulant. In our case, the higher release of carvacrol and be considered the difference in polarity between the polar migrated
thymol into 950 mL/L ethanol rather than 100 mL/L ethanol showed substances and the non-polar polymer.
the influence of simulant polarity and AO solubility. Also, it can be
assumed that certain amount of simulant will penetrate into the 3.3. Antioxidant activity of migration extracts
matrix, enhancing the mobility of the target AOs inside the polymer
chains, which could promote migration. This behaviour has also The antioxidant activity of carvacrol and thymol has been re-
been suggested in previous studies for the migration of some AOs ported in previous studies, although the mechanism of such activity
from polyolefins into fatty food simulants (Haider & Karlsson, is not fully understood (Yanishlieva, Marinova, Gordon, & Raneva,
2000; Kuorwel et al., 2013; Peltzer et al., 2009; Tovar et al., 2005). 1999). The antioxidant activity depends not only on the com-
Regarding aqueous food simulants, migration of AOs was also pound structure but also on many other factors, such as concen-
observed although the solubility of both compounds in aqueous tration, temperature, light, simulant type and physical state of the
solutions is low; with migration values increasing by using acetic system (e.g. pH).
acid 30 g/L and ethanol 100 mL/L. The migration of AOs into these The antioxidant capacities of the obtained extracts were eval-
simulants might be due mainly to two factors: the hydrophilic uated by DPPH radical assay (Fig. 1). No DPPH inhibition was
character of these additives described in literature (Peltzer et al., observed in acetic acid 30 g/L, possibly due to the pH of this sim-
2009); and the small size of these compounds, and thus faster ulant. The absorbance of DPPH could decrease by light exposure,
diffusion, since the diffusion rate is governed by the mobility of the oxygen content, pH, and solvent type (Ozcelik et al., 2003).
additives which is determined by the size and geometry of the Regarding pH, it has been reported that generally, the increase of

Table 3
Release of AOs (mg/kg simulant) obtained from PP films into aqueous and fatty food simulants under conditions according to European Standard EN 13130-2005 (n ¼ 3,
m  SD).

Analyte Film Simulant

Watera Ethanol 100 mL/La Acetic acid 30 g/La Ethanol 950 mL/La Isooctaneb

Carvacrol PPC8 288  20 718  54 647  47 880  27 921  157

PPTC8 157  19 285  26 474  44 347  21 633  34
Thymol PPT8 433  46 656  30 689  61 829  19 1085  112
PPTC8 162  18 362  53 547  51 367  21 616  49

Migration conditions.
a
40  C, 10 days.
b
20  C, 2 days.
474 M. Ramos et al. / LWT - Food Science and Technology 58 (2014) 470e477

hydrogen ion concentration leads to the decrease of the reaction performance is the kinetics of release of the AO agent from the
rate of chromogen radical scavenging, whereas under basic condi- packaging. In order to have a deeper knowledge of the migration
tions proton dissociation of polyphenolics would enhance the mechanism of the target AOs from PP films, kinetic experiments
reducing capacity of compounds (Pyrzynska & Pekal, 2013). On the were carried out by using four different food simulants during 15
other hand, all the other extracts presented an appreciable anti- days. For this study, only formulations containing the studied AOs
oxidant activity, showing a significant inhibition of the DPPH at 80 g/kg separately were considered.
radical. A higher antioxidant capacity was observed for thymol Figs. 3 and 4 show the release of thymol (a) and carvacrol (b)
extracts, with the highest inhibition obtained into isooctane from PPT8 and PPC8 films as a function of time, respectively. Similar
(42.2  1.1%). The ANOVA results also showed that independently of behaviour was observed for both compounds, being rapidly
the variations introduced by the use of the different simulants, the released from their respective films into the studied food simulants,
formulation with 80 g/kg of thymol was significantly different from with an expected increase in their release with increasing time and
the other ones regarding antioxidant capacity (Fig. 2). reaching equilibrium after approximately 120 h.
These results showed that thymol antioxidant activity was su- In a first approach, the higher amount of migrated analytes
perior to that of carvacrol, possibly due to greater steric hindrace of occurs to isooctane (see a values in Table 4). Furthermore, for this
the thymol phenolic group; as different authors have concluded simulant, at t > 120 h, the equilibrium is not well defined and, for
when considering the mechanism of action of these compounds in example, for PPT8 almost all previously encapsulated thymol is
the DPPH assay (Wu, Luo, & Wang, 2012; Yanishlieva et al., 1999). released in isooctane. This can be argued in two different ways:
Other compounds having a hydroxyl group sterically hindered, either by experimental or physical grounds. Starting with the
such as BHT, have been also reported to possess high antioxidant former, it seems that the quantification of AOs in isooctane, for
activity (Mastelic et al., 2008). longer times, lead to scatter values and, consequently, the some-
The DPPH inhibition values were correlated with the AOs what relevant increase of the amount of AOs released after 15 days
amount released from the films (Table 3). The highest amount of cannot be relevant from statistics point of view. Another possible
released additives was observed into fatty food simulants; although justification is based on the behaviour of the release of active
no significant differences between isooctane and ethanol 950 mL/L compounds from polyolefin films immersed in food simulants by
were observed with different formulations at P < 0.05 (Fig. 1). The the “swelling-controlled” model (Suppakul et al., 2011). According
obtained results indicate that a considerable quantity of the AOs to this model, a simulant penetrates first into the polymer matrix
remain in the polymer matrix and consequently could act as active and dissolves the active agents thereby enabling their subsequent
agents in these materials. In this sense, the obtained PP films could release. Indeed, it is expected that a simulant uptake will cause
be used as AO films for food packaging applications in order to polymer swelling. The migration of AOs is thus expected to increase
extend the shelf-life of food products, retarding oxidation pro- with an increase in the simulant penetration into the PP film,
cesses. In addition, it has been reported that these additives could reaching a plateau when the matrix is saturated with the simulant.
be also used to protect the polymer against oxidative degradation However, many interactions take place during the migration of
during processing and further use (Ramos et al., 2012). species from polymers into liquids. Moreover, it has pointed out
Finally, the study of the combined activity of carvacrol and that a time-dependent relaxation process could occur as a result of
thymol in the same film at 40 g/kg of each compound (PPTC8) the swelling that takes place during the diffusion of the liquid into
(Fig. 1) showed some additive effect between them as similar re- the polymer. As a consequence, release rates change continuously
sults were obtained for samples with 80 g/kg of each compound and the accurate mathematical analysis of the migration is difficult.
(PPC8 and PPT8) separately. This effect was more evident into fatty The penetration of simulant molecules facilitates further penetra-
food simulants. tion by the plasticization of the polymer matrix, until a plateau is
reached. As pointed out before, for isooctane an increase of
migration after reaching the equilibrium was observed for both
3.4. Kinetics of AOs migration from active films
studied AOs at 360 h. This could be due to a combination of tem-
perature and a longer time in which the PP films were penetrated
Information about diffusion coefficients through packaging
by isooctane producing the increase on the release. Also, it can be
materials is, in general, very useful to evaluate the performance of
speculated about the sorption of isooctane by the PP matrix and a
new active packaging materials, as the critical point in antioxidant
consequent creation of void spaces favouring the migration of the
phenolic compounds (Manzanarez-López et al., 2011).
The experimental release data shown in Fig. 3 were further
analysed in terms of a diffusion model according to Equation (3).
However, the use of this equation in order to compute diffusion
coefficients needs the previous knowledge of partition values. Ac-
cording to the previous discussion it can be assumed that the
amount of AO release can be estimated as being constant after ca.
120 h, and once Equation (3) can only be applied to MF,t/MP,0 < 0.6,
the use of this equation do not interfere with the hypothetical (not
confirmed) effect of the swelling-process in the mass transport by
diffusion. In these circumstances the release of AOs from PPT8 and
PPC8 to different simulations can be modelled by using the
following equation:

    0

MF;t MP;0 ¼ MF;N MP;0 1  ek t (5)

Fig. 2. Plot representing mean DPPH inhibition values (%) for the formulations PPT8, where k0 is a constant related with the release rate constant. By
PPC8 and PPTC8. Different letter represent significant differences at P < 0.05. fitting Eq. (5) to experimental AOs release data (see solid lines in
 M. Ramos et al. / LWT - Food Science and Technology 58 (2014) 470e477 475

Fig. 3. Release of thymol from PPT8 into different food simulants over 15 days. (A) Ethanol 100 mL/L, 40  C; (B) acetic acid, 40  C; (C) Ethanol 950 mL/L, 40  C; and (D) isooctane,
20  C. Solid lines were obtained by fitting Eq. (4) to experimental data. For further details see Section 3.4.

Fig. 4. Release of carvacrol from PPC8 into different food simulants over 15 days. (A) Ethanol 100 mL/L, 40  C; (B) acetic acid, 40  C; (C) Ethanol 950 mL/L, 40  C; and (D) isooctane,
20  C. Solid lines were obtained by fitting Eq. (4) to experimental data. For further details see Section 3.4.
476 M. Ramos et al. / LWT - Food Science and Technology 58 (2014) 470e477

Table 4
Diffusion coefficients (D
1014, m2/s) calculated from Equation (3) for the release of AOs from PP films into different food simulants (m  SD, n ¼ 3).

Analyte (film) Simulant

Ethanol 100 mL/La Acetic acid 30 g/La Ethanol 950 mL/La Isooctaneb

Carvacrol (PPC8) aap 1.38  0.07 0.96  0.04 1.27  0.07 1.4  0.1
D 1.20  0.05 1.7  0.1 1.99  0.07 9.4  0.6
Thymol (PPT8) aap 1.56  0.1 1.44  0.05 1.56  0.08 2.4  0.2
D 1.75  0.08 2.51  0.02 1.01  0.03 5.9  0.1

Migration temperature.
a
40  C.
b
20  C.

Figs. 2 and 3), the values of MF,N/MP,0 can be obtained and, finally Guilherme, Rubira, & Muniz, 2007). In general, the fitting deter-
apparent partition coefficients (aap) values can be calculated mination coefficients are higher than 0.96, with the exception of
through the equation: isooctane-containing systems where determination coefficients of
0.819 (PPC8) and 0.713 (PPT8) were obtained.
   The results obtained for thymol and carvacrol are shown in
MFN MP;0 ¼ aap 1 þ aap (6) hFig. 5A and i0:5 B, respectively. As it can be seen, the linearity of
1 1 MF;t versus t0.5 was very good for both AOs and all simu-
Both parameters, MF,N/MP,0 and a are reported in Table 4. p  a$MF;0
lants tested with determination coefficient values (R2) ranging from
It should be stressed that the choice of Eq. (5) has been done
0.961 to 0.995 for thymol and 0.983e0.992 for carvacrol, suggesting
once, for the border and limit conditions of the experiments re-
that experimental release data is well described by the proposed
ported in this work, it describes a first order kinetic process (Reis,
diffusion model for short-range times.
The analysis of diffusion coefficients (D) (Table 4) shows that the
diffusion process in different simulants are independent on the
AOs, with D values ranging from 1
1014 to 2
1014 m2/s. This
behaviour was expected if considering that carvacrol and thymol
are isomers having similar molecular weights and polarity
(Licciardello, Muratore, Mercea, Tosa, & Nerin, 2013). The exception
occurs for the diffusion of AOs into isooctane. In fact, D values for
thymol and carvacrol are 4 and 6 times higher for this simulant
than the average values for remaining ones. This is, however, in line
with the discussion carried out in the previous section and with
results reported in Section 3.2 as well. It is also worth noticing that
the magnitude of D values found for these films are one order of
magnitude lower than those obtained for similar AOs and films
(Suppakul et al., 2011), suggesting that these films can provide a
long term release, or higher retention inside films, of AOs.

4. Conclusions

The release study of carvacrol and thymol from PP films into

aqueous and fatty food simulants was accomplished. Analytical
methods for the determination of the target compounds in the
studied food simulants were successfully developed and validated.
Release of AOs from PP films showed some differences depending
on the food simulant type used; being isooctane the most
exhaustive one resulting in high levels of migration. In addition,
positive results were obtained for all migration extracts by the
antioxidant activity study performed by the DPPH method,
showing thymol a higher antioxidant capacity. Finally, the results
obtained for the migration kinetics study showed that carvacrol
and thymol incorporated into PP films at 80 g/kg were readily
released into different food simulants, being these additives still
remaining in the polymer after 15 days. In this sense, the high ef-
ficiencies of release of these compounds from PP films point to the
great potential of these systems in antioxidant packaging of
different food products to extend their shelf life and avoid the direct
addition of additives to food formulations.

h i Acknowledgements
MF;t 0:5
Fig. 5. Plots of p1  1a$MF;0  p10:5 versus t0.5 for the migration of thymol (A) and
carvacrol (B) from PPT8 and PPC8 films, respectively, into different food simulants.
Isooctane (>), 20  C; acetic acid (B), 40  C; ethanol 100 mL/L (,), 40  C; and ethanol Authors would like to thank Ashland Chemical Hispania for
950 mL/L (6), 40  C. kindly supplying ECOLEN HZ10K PP (Hellenic Petroleum) and
 M. Ramos et al. / LWT - Food Science and Technology 58 (2014) 470e477 477

Spanish Ministry of Economy and Competitiveness for financial Ozcelik, B., Lee, J. H., & Min, D. B. (2003). Effects of light, oxygen, and pH on the
absorbance of 2,2-Diphenyl-1-picrylhydrazyl. Journal of Food Science, 68(2),
support (MAT-2011-28648-C02-01). Also, Marina Ramos would like
487e490.
to thank University of Alicante (Spain) for UAFPU2011-48539721S Park, H.-Y., Kim, S.-J., Kim, K. M., You, Y.-S., Kim, S. Y., & Han, J. (2012). Development
predoctoral research grant. of antioxidant packaging material by applying corn-zein to LLDPE film in
combination with phenolic compounds. Journal of Food Science, 77(10), E273e
E279.
References Peltzer, M., Wagner, J., & Jiménez, A. (2009). Migration study of carvacrol as a
natural antioxidant in high-density polyethylene for active packaging. Food
Barbara, G., & Tabachnick, L. S. F. (2013). Using multivariate statistics: Pearson new Additives & Contaminants: Part A: Chemistry, Analysis, Control, Exposure & Risk
international edition (6th ed). Pearson Education. Assessment, 26(6), 938e946.
Barbosa-Pereira, L., Cruz, J. M., Sendón, R., Rodríguez Bernaldo de Quirós, A., Ares, A., Pyrzynska, K., & Pekal, A. (2013). Application of free radical diphenylpicrylhydrazyl
& Castro-López, M. (2013). Development of antioxidant active films containing (DPPH) to estimate the antioxidant capacity of food samples. Analytical
tocopherols to extend the shelf life of fish. Food Control, 31(1), 236e243. Methods, 5(17), 4288e4295.
Burman, L., Albertsson, A.-C., & Höglund, A. (2005). Solid-phase microextraction for Ramos, M., Jiménez, A., Peltzer, M., & Garrigós, M. C. (2012). Characterization and
qualitative and quantitative determination of migrated degradation products of antimicrobial activity studies of polypropylene films with carvacrol and thymol
antioxidants in an organic aqueous solution. Journal of Chromatography A, for active packaging. Journal of Food Engineering, 109(3), 513e519.
1080(2), 107e116. Reis, A. V., Guilherme, M. R., Rubira, A. F., & Muniz, E. C. (2007). Mathematical model
Byun, Y., Kim, Y. T., & Whiteside, S. (2010). Characterization of an antioxidant pol- for the prediction of the overall profile of in vitro solute release from polymer
ylactic acid (PLA) film prepared with a-tocopherol, BHT and polyethylene glycol networks. Journal of Colloid and Interface Science, 310(1), 128e135.
using film cast extruder. Journal of Food Engineering, 100(2), 239e244. Reynier, A., Dole, P., Humbel, S., & Figenbaum, A. (2001). Diffusion coefficients of
Crank, J. (1975). The mathematics of diffusion (2nd ed.). Oxford: Oxford University Press. additives in polymers. I Correlation with geometric parameters. Journal of
Chung, D., Papadakis, S. E., & Yam, K. L. (2002). Simple models for assessing migration Applied Polymer Science, 82, 2422e2433.
from food-packaging films. Food Additives and Contaminants, 19(6), 611e617. Ridgway, K., Lalljie, S. P. D., & Smith, R. M. (2007). Sample preparation techniques for
Del Nobile, M. A., Conte, A., Buonocore, G. G., Incoronato, A. L., Massaro, A., & the determination of trace residues and contaminants in foods. Journal of
Panza, O. (2009). Active packaging by extrusion processing of recyclable and Chromatography A, 1153(1e2), 36e53.
biodegradable polymers. Journal of Food Engineering, 93(1), 1e6. Salafranca, J., Pezo, D., & Nerín, C. (2009). Assessment of specific migration to
Dopico-García, M. S., Castro-López, M. M., López-Vilariño, J. M., González- aqueous simulants of a new active food packaging containing essential oils by
Rodríguez, M. V., Valentão, P., & Andrade, P. B. (2011). Natural extracts as po- means of an automatic multiple dynamic hollow fibre liquid phase micro-
tential source of antioxidants to stabilize polyolefins. Journal of Applied Polymer extraction system. Journal of Chromatography A, 1216(18), 3731e3739.
Science, 119(6), 3553e3559. Siró, I., Fenyvesi, É., Szente, L., De Meulenaer, B., Devlieghere, F., & Orgoványi, J.
EC. (2011). Commission regulation (EU) no 10/2011 of 14 January 2011 on plastic (2006). Release of alpha-tocopherol from antioxidative low-density poly-
materials and articles intended to come into contact with food. Official Journal ethylene film into fatty food simulant: influence of complexation in beta-
of the European Union, 2e89. cyclodextrin. Food Additives and Contaminants, 23(8), 845e853.
Haider, N., & Karlsson, S. (2000). Kinetics of migration of antioxidants from poly- Suppakul, P., Sonneveld, K., Bigger, S. W., & Miltz, J. (2011). Diffusion of linalool and
olefins in natural environments as a basis for bioconversion studies. Bio- methylchavicol from polyethylene-based antimicrobial packaging films. LWT e
macromolecules, 1, 481e487. Food Science and Technology, 44(9), 1888e1893.
Kuorwel, K. K., Cran, M. J., Sonneveld, K., Miltz, J., & Bigger, S. W. (2013). Migration of Tehrany, E. A., Mouawad, C., & Desobry, S. (2007). Determination of partition co-
antimicrobial agents from starch-based films into a food simulant. LWT e Food efficient of migrants in food simulants by the PRV method. Food Chemistry,
Science and Technology, 50(2), 432e438. 105(4), 1571e1577.
Licciardello, F., Muratore, G., Mercea, P., Tosa, V., & Nerin, C. (2013). Diffusional Tomaino, A., Cimino, F., Zimbalatti, V., Venuti, V., Sulfaro, V., & De Pasquale, A.
behaviour of essential oil components in active packaging polypropylene films (2005). Influence of heating on antioxidant activity and the chemical compo-
by multiple headspace solid phase microextractionegas chromatography. sition of some spice essential oils. Food Chemistry, 89(4), 549e554.
Packaging Technology and Science, 26(3), 173e185. Tovar, L., Salafranca, J., Sanchez, C., & Nerin, C. (2005). Migration studies to assess
Lopez-de-Dicastillo, C., Nerin, C., Alfaro, P., Catala, R., Gavara, R., & Hernandez- the safety in use of a new antioxidant active packaging. Journal of Agricultural
Munoz, P. (2011). Development of new antioxidant active packaging films based and Food Chemistry, 53(13), 5270e5275.
on Ethylene Vinyl Alcohol Copolymer (EVOH) and green tea extract. Journal of UNE-EN. (2005). UNE-EN 13130-1:2005. Materials and articles in contact with food-
Agricultural and Food Chemistry, 59(14), 7832e7840. stuffs. Plastics substances subject to limitation. Guide to test methods for the spe-
Manzanarez-López, F., Soto-Valdez, H., Auras, R., & Peralta, E. (2011). Release of cific migration of substances from plastics to foods and food simulants and the
[alpha]-Tocopherol from Poly(lactic acid) films, and its effect on the oxidative determination of substances in plastics and the selection of conditions of exposure
stability of soybean oil. Journal of Food Engineering, 104(4), 508e517. to food simulants.
Mastelic, J., Jerkovic, I., Blazevic, I., Poljak-Blazi, M., Borovic, S., & Ivancic-Bace, I. Wu, Y., Luo, Y., & Wang, Q. (2012). Antioxidant and antimicrobial properties of
(2008). Comparative study on the antioxidant and biological activities of essential oils encapsulated in zein nanoparticles prepared by liquideliquid
carvacrol, thymol, and eugenol derivatives. Journal of Agricultural and Food dispersion method. LWT e Food Science and Technology, 48(2), 283e290.
Chemistry, 56(11), 3989e3996. Yanishlieva, N. V., Marinova, E. M., Gordon, M. H., & Raneva, V. G. (1999). Antioxi-
Mastromatteo, M., Barbuzzi, G., Conte, A., & Del Nobile, M. A. (2009). Controlled dant activity and mechanism of action of thymol and carvacrol in two lipid
release of thymol from zein based film. Innovative Food Science & Emerging systems. Food Chemistry, 64(1), 59e66.
Technologies, 10(2), 222e227.