Migration of toluene through different plastic laminated films into food simulants

Food Control 59 (2016) 164e171

Contents lists available at ScienceDirect

Food Control journal homepage: www.elsevier.com/locate/foodcont

Migration of toluene through different plastic laminated films into food simulants Nan Chang a, b, Chun-hong Zhang a, Feng-e Zheng b, Ya-lu Huang a, Jin-yan Zhu a, c, Qian Zhou a, Xin Zhou a, Shu-juan Ji a, * a b c

Department of Food Science, Shenyang Agricultural University, Shenyang, 110866, PR China Food Safety Institute, Shenyang Product Quality Supervision and Inspection Institute, Shenyang, 110022, PR China Zhuanghe Food Inspection and Supervision Center, Dalian, 116400, PR China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 26 December 2014 Received in revised form 22 April 2015 Accepted 26 April 2015 Available online 19 May 2015

Laminated films are generally prepared from two or more monolayer films bonded using adhesives, such that the printing ink and adhesives are located between the films. Although the organic solvents in the adhesive, printing ink, and diluter are not in direct contact with the food, they may migrate through the films under certain conditions, potentially threatening human health. In this study, gas chromatography was used to analyze the migration rate of residual toluene from laminated films into four food simulants, namely isooctane, 50% ethanol, 3% acetic acid, and 10% ethanol. The effect of different parameters on toluene migration was characterized for internal films prepared with either low density polyethylene (LDPE) or cast polypropylene (CPP). The migration rate varied markedly with the temperature, with higher temperatures leading to accelerated migration into the simulants. The toluene levels in the simulants plateaued after a certain time period. Isooctane was the simulant into which toluene migrated the fastest. CPP films proved more efficient barriers than those prepared with LDPE. Within a certain range, increasing the initial toluene content in the laminated films had almost no observable effect on its migration rate. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Toluene Migration Laminated films Food simulants Food packaging

1. Introduction Plastic films are mechanically strong and chemically stable and thereby play an important role in food packaging. Laminated films are prepared from two or more films bonded using adhesives, so that the beneficial properties of both may be utilized simultaneously and provide an optimal and flexible packaging material for food preservation. These laminated films significantly increase the shelf life of various packaged foods such as fish, meat, cheese, sausage, poultry, beverages, and sauerkraut. Low density polyethylene (LDPE) and cast polypropylene (CPP) films form effective moisture barriers and provide good heat sealing. In laminated films (Twede & Goddard, 1998), they are widely used for the inner layer, which is in direct contact with the food. In industry, plastic composite films are prepared either by extrusion or dry lamination (Torres, Guarda, Moraga, Romero, & Galotto,

* Corresponding author. E-mail addresses: [email protected] (N. Chang), [email protected] (S.-j. Ji). http://dx.doi.org/10.1016/j.foodcont.2015.04.042 0956-7135/© 2015 Elsevier Ltd. All rights reserved.

2012), with the latter being the simpler, less expensive, and therefore most widely used process, accounting for more than 70% of all composite film production. In dry lamination, a single-layer film is coated with an adhesive that is then evaporated in a drying oven. The coated film is then laminated onto another singlelayer film at high temperature and pressure. The substances used for printed lamination filmsdprinting inks, adhesives, and organic solventsdwhich are located between the two layers, may nonetheless migrate through the latter and into the foodstuffs (Dong, Li, & Zhang, 2011; Katan, 1996; Reinas, Oliveira, Pereira, Machado, & Pocas, 2012). The two solvents most often used for the ink and adhesives are toluene and ethyl acetate. Toluene is one of the most widely used aromatic solvents but is a known carcinogen and teratogen (Bowen & Hannigan, 2013). It is toxic furthermore to the liver and to the hematological and immune systems. The migration of solvents such as toluene from wrapping into food therefore affects not only the taste and quality of the latter, but is also a potential health hazard. The diffusion of chemical substances through plastic films is a very complex process that depends on several factors such as the

N. Chang et al. / Food Control 59 (2016) 164e171

concentration of the substances in the packaging and the food, the fat content of the latter, the temperature, and the storage time (Bhuni, Sablani, Tang, & Rasco, 2013). The migration process can be divided into four major steps: the sorption of the compounds at the plasticefood interface, the diffusion of chemical compounds through the polymers, the desorption of the diffused molecules from the polymer surface, and the diffusion of the compounds in the food (Coltro et al., 2014; Ferrara, Bertolodo, Scoponi, & Ciardelli, 2001). Previous studies have highlighted the presence of many compounds in the different laminated films on the market, 66% of which migrated into the simulant Tenax (a solid food simulants), with toluene in particular reaching 1.2 mg/kg (Vera, Canellas, & Nerín, 2014). A study of the migration of alkylbenzenes from offset printing inks into hamburger rolls found 2 mg/kg alkylben€ derhjelm, 2001). Goulas zenes therein. (Aurela, Ohra-aho, & So (2001) showed that the overall migration from five-layer coextruded packaging films into aqueous food simulants was 2.3e15.9 mg/L. Elsewhere, Meng et al. have characterized the risk of food contamination arising from the migration of alkylbenzenes from paper and plastic packaging materials into food (Meng, Liao, Sun, Liao, & Liu, 2007). According to European Union (EU) regulations, the quality of food packaging materials is assessed mainly via two migration indexes: the overall migration limit (OML) and the specific migration limit (SML). The regulations state that plastic packages should not transfer their constituents to food simulants in quantities exceeding the OML, namely 10 mg in total per dm2 of food contact surface (mg/dm2). Furthermore, the contents in the food of particular packaging constituents should be below 60 mg/kg, the SML (Reg 10/2011). Furthermore, all food-contact materials need to follow European Commission Regulation 1935/2004 (Reg 1935/ 2004), which states that substances migrating into food should not be harmful to humans. Furthermore, benzene derivatives are strictly prohibited in foodstuffs, both in Europe and the United States. The EU has no specific regulations concerning adhesives or ink in plastic laminated films. Nonetheless, such limits have been included in the corporate standards of certain companies. For example, Coexpan ensures that the total amount of solvent residues transferred into its wrapped foodstuffs does not exceed 20 mg/m2 (Mao, Zheng, Yu, Jiang, & Chen, 2008). In the People's Republic of China (PRC), the General Administration of Quality Supervision, Inspection and Quarantine has published Review Guidelines for the Manufacturing License of Food-Grade Plastic Package, Container, Tools and Other Products, which states that the total amount of solvent residues in plastic laminated film (pouch) products should be less than 10 mg/m2 in general, and less than 2 mg/m2 for benzene derivatives (AQSIQ, 2006). The national standard GB/T 10004-2008 specifies the acceptable level of solvent residues in food-grade laminated films. The total solvent content should be less than 5.0 mg/m2 and no benzene derivatives should be detected (AQSIQ, 2008). However, this standard only considers three types of laminated film whereas more than 20 of these are currently in use on the market, with their own specific company standards that may differ in terms of their residual solvent limits. Indeed, rather than banning it completely, some companies only require benzene derivative contents to be less than 2 mg/m2. Clearly, although efforts are being made in the PRC to resolve problems related to solvents in food-grade laminated films, the many factors that can affect solvent migration into food mean that systematic studies thereof are currently lacking. In this study, two plastic laminated films prepared with different inner layers were chosen to study the effects, among others, of temperature, simulants, time and inner layer materials on migration. In view of reducing the hazards associated with toluene

165

ingestion, the results presented herein provide a scientific basis for future regulations concerning the migration of toluene from laminated films into foodstuffs. 2. Materials and methods 2.1. Reagents Chromatography grade toluene, isooctane, anhydrous ethanol, and acetic acid were purchased from Sinopharm Chemical Reagent Co., Ltd, Shenyang, PRC. Ultrapure water was prepared using a MilliQ filtration system. According to relevant EU rules and regulations, as well as for practical reasons, migration tests were conducted on the films in contact with the following four food simulants (Reg 10/2011; Directives 82/711/EEC & 85/572/EEC). Simulant A: isooctane; Simulant B: ethanol 50% (v/v) aqueous solution (Milli-Q water); Simulant C: ethanol 10% (v/v) aqueous solution (Milli-Q water); Simulant D: acetic acid 3% (w/v) aqueous solution (Milli-Q water). 2.2. Samples Two compositions were used for the printed laminate samples, namely biaxial oriented propylene (BOPP)/LDPE and BOPP/CPP, with BOPP forming the outer layer and LDPE or CPP the inner one in direct contact with the food. The samples were provided by Shenyang Bafang Plastic Packaging Co., Ltd. The laminated films were heat-sealed into pouches. In this study, four types of laminated film samples were prepared, namely: LDPE30: a BOPP/LDPE bilayer film with a 30 mm thick LDPE (inner) layer; LDPE50: a BOPP/LDPE bilayer film with a 50 mm thick LDPE layer; CPP30: a BOPP/CPP bilayer film with a 30 mm thick CPP layer; CPP50: a BOPP/CPP bilayer film with a 50 mm thick CPP layer. The effect of the initial toluene concentration on its migration rate through different laminated films was investigated. The initial toluene contents of samples in groups A, B, and C were 1.371e1.402 mg/m2, 2.407e2.593 mg/m2, and 3.612e3.708 mg/m2, respectively (Table 1). 2.3. Migration tests According to EU regulation No 10/2011, isooctane, 50% ethanol, 10% ethanol and 3% acetic acid were selected as food simulants with a surface-to-volume ratio of 1 dm2/20 mL (Reg 10/2011), and single-surface migration tests were performed within the pouches. The food simulants were packed into the pouches and then

Table 1 Initial toluene contents of the different laminated films (n ¼ 3). Sample

Initial content ± SD (mg/m2)

LDPE30 LDPE50 CPP30 CPP50

1.373 1.402 1.371 1.381

A

B ± ± ± ±

0.021 0.010 0.017 0.014

2.561 2.593 2.407 2.586

C ± ± ± ±

0.029 0.037 0.028 0.030

3.612 3.708 3.669 3.645

± ± ± ±

0.025 0.036 0.021 0.033

166

N. Chang et al. / Food Control 59 (2016) 164e171

vacuum-heat-sealed to ensure adequate contact between the liquid simulants and the composite pouches (AQSIQ, 2009). 2.3.1. Migration tests under different temperatures The samples prepared as described above were placed under 4  C, 10  C, 20  C, 30  C, 40  C, and 50  C for migration tests. After 10 d, the toluene migration rate was calculated as follows (CEN, 2004):

M¼

C1  100; C2

(1)

where M (%) is the migration rate, and C1 and C2 (mg/m2) are the toluene concentrations respectively in the simulant after migration and in the laminated film, initially. 2.3.2. Migration tests at different time intervals According to EU regulation No 10/2011, the migration temperature was set to 40  C (Reg 10/2011), and the toluene migration rate was measured at different time intervals (1, 5, and 10 h, and 1, 2, 3, 4, 5, 6, 7, 8, 10, and 12 d). 2.4. Instrumental analysis 2.4.1. Method for measuring the initial toluene content of the samples The samples were pretreated according to the GB/T 10004-2008 standard (AQSIQ, 2008), and their initial toluene contents were measured as proposed by Xu et al. (Xu, Yang, Gao, & Lu, 2008). 2.4.2. Method for measuring toluene migration levels in the simulants Toluene samples were weighed precisely (1 g) and mixed with isooctane to a volume of 100 mL. After adequate shaking, the mixture was conserved as a parent stock solution and diluted with isooctane to obtain six standard solutions with concentrations of 0.5 mg/mL, 0.3 mg/mL, 0.1 mg/mL, 0.05 mg/mL, 0.02 mg/mL, and 0.01 mg/mL, which were immediately transferred into a refrigerator. Prior to gas chromatography (GC) experiments, the standard solutions were separately added to 50 mL preheated headspace vials, sealed, and heated for 30 min at 100  C. After migration, the liquid simulants were concentrated using a Turbo VaP II workstation (Caliper Life Sciences Co.), and isooctane was added up to a volume of 1 mL. Again, before GC, 50 mL of the different mixtures were added to preheated vials, sealed, and heated for 30 min at 100  C. The toluene contents were calculated based on the standard GC curves. Each experiment, including all the steps described above, was repeated three times to obtain a mean migration rate for each simulant. 2.4.3. Gas chromatography parameters The GC system used in this study was an Agilent 6890N with a flame ionization detector and a head-space sampler. Separation was achieved with an Agilent DB-624 (30 m  0.25 mm, 1.40 mm) column; the nitrogen flow rate was 1.0 mL/min, the injector and

detector temperatures were 220  C and 250  C, respectively, and split injection was used with an injection volume of 1 mL and a split ratio of 10:1. The temperature was initially set to 45  C (6 min), then increased at 20  C/min up to 130  C and maintained for 2 min. 2.4.4. Infrared spectroscopy analysis The infrared spectrometer used in this study was an FT-IR200 manufactured by Thermo Fisher Scientific (China) Co., Ltd. To eliminate the effects of the outer layer of the laminated films on the infrared spectroscopy measurements, background measurements were obtained from the food simulants placed in LDPE and CPP monolayer pouches and treated as described above. 3. Results 3.1. Quality control analysis The toluene contents of the films and the simulants were obtained as described in Sections 2.4.1 and 2.4.2, and by analyzing the resulting data by linear regression. The curves were found to be linear for toluene concentrations between 0.027 mg/m2 and 18.5 mg/m2 and between 0.027 mg/m2 and 13.9 mg/m2 in the films and simulants, respectively. The limit of detection (LOD) and limit of quantitation (LOQ) were calculated with signal-to-noise ratios (S/N) of three and ten, respectively, corresponding to toluene contents of 0.008 mg/m2 and 0.027 mg/m2 (Table 2). Toluene solutions at either 1.0 mg/L, 0.5 mg/L and 0.1 mg/L were added to the simulant solutions before the mixture was concentrated, adjusted to a certain volume as described in Section 2.4.2, and analyzed by GC. Recovery was calculated based on the ratio of the calculated concentration and the added concentration. The residual standard deviation (RSD) of five replicate measurements was also calculated (Table 3). The recovery and RSD for toluene were 85.2e98.7e% and 3.5e7.7%, respectively, indicating a sufficient recovery and precision for toluene detection in all four simulants (Table 3). 3.2. Migration tests 3.2.1. Temperature dependence of the toluene migration rate The migration of toluene into food simulants through CPP and LDPE films was quantified at different temperatures. Considering the conditions under which common laminated films are used in practice, six temperature points between 4  C and 50  C were chosen. The migration rates of toluene at typical refrigeration temperatures (4  C) were found to be low through each of the four films and into each of the four simulants (Fig. 1AeD). As shown in Fig. 1A, the migration rates of toluene through LDPE30 and LDPE50 into isooctane are different, but both increase significantly with the temperature (P < 0.01). The migrations rates through these two films are respectively 44.3 and 46.3 percentage points higher at 50  C than at 4  C. For the CPP30 and CPP50 films, the increase in the toluene migration rate is only slight up to 10  C, but more significant thereafter (P < 0.05). The toluene migration

Table 2 Quality control analysis (n ¼ 3). Method

Detection of toluene in the sample Detection of toluene in the simulants a b

LOD ¼ 3  S/N. LOQ ¼ 10  S/N.

Quality control analysis Equation

R2

Linear range(mg/m2)

LODa(mg/m2)

LOQb (mg/m2)

y ¼ 3075.3 x0.3537 y ¼ 3299.3 x0.5331

0.9999 0.9998

0.027e18.5 0.027e13.9

0.008 0.008

0.027 0.027

N. Chang et al. / Food Control 59 (2016) 164e171 Table 3 Recoveries (%) and RSDa (%) of toluene in the different food simulants (n ¼ 5). Simulants

isooctane 50% ethanol 10% ethanol 3% acetic acid a

Recovery, % (RSD, %) 1.0 mg/L

0.5 mg/L

0.1 mg/L

95.3 90.6 87.9 92.5

98.7 92.8 85.9 87.4

96.2 87.6 91.5 85.2

(3.7) (5.2) (4.7) (5.6)

(6.0) (6.6) (3.5) (5.6)

(4.9) (6.4) (6.9) (7.7)

RSD: Relative standard deviation of 5 experimental determinations.

rate through CPP50 increases with the temperature, in particular from 10 to 20  C and 30e40  C (P < 0.05, Fig. 1A). For the 50% ethanol simulant, a significant increase in the toluene migration rate is observed through the LDPE30 and LDPE50 films with increasing temperatures (P < 0.05) (Fig. 1B). For CPP30 in contrast, the migration rate varies little between 10  C and 30  C (P > 0.05), and a significant increase is only observed for temperatures greater than 30  C (P < 0.05) (Fig. 1B). The toluene migration rate also increases with the temperature, but not significantly beyond 40  C. The increase in the toluene migration rate into 10% ethanol is clear for the LDPE and CPP30 samples, for temperatures higher than 20  C and up to 40  C, respectively (P < 0.05) (Fig. 1C). In contrast, for temperatures higher than 30  C, the migration rate through the CPP50 film does not vary significantly within the experimental temperature range (P > 0.05). With the 3% acetic acid simulant, the migration rate significant increases with the temperature through the LDPE30 and LDPE50 films (P < 0.05) (Fig. 1D), more slowly though the CPP30 one, and through the CPP50 film, an significant increase is only apparent for temperatures above 20  C (P < 0.05) (Fig. 1D). 3.2.2. Time dependence of the toluene migration rate On the whole, the toluene migration rates through each film type into the four simulants tested all tend to increase over time, plateauing after a certain delay (Fig. 2 & Table 4). The result shows that the film sample with the shortest equilibration time is LDPE30, for which the toluene migration rate into isooctane reaches 76.27%

167

after 48 h. The migration rate is slower to equilibrate with the CPP50 film, taking 192 h to reach 13.48% in the 10% ethanol simulant solution. In terms of the simulants, the migration rates equilibrate most rapidly for isooctane and 50% ethanol, with a particularly sharp increase in the first 24 h. In contrast, for 3% acetic acid and 10% ethanol, the migration rate increases slowly over time, taking 72e192 h to reach equilibrium. One should note furthermore that no toluene is detected in the 10% ethanol simulant before 1 h and 10 h with the LDPE and CPP films, respectively. Similarly, the toluene in the 3% acetic acid simulant remains below detection levels up to 5 h and 10 h, respectively, for the CPP30 and CPP50 samples. 3.2.3. Simulant dependence of the toluene migration rate The migration experiments were conducted at 40  C for 10 d, as specified in the relevant standards, laws, and regulations. Isooctane is the simulant in which the highest toluene migration rates are recorded, 49.95e80.25% depending on the film (Fig. 3). In 50% ethanol, the migrations rates range from 27.89% to 69.13%. With the LDPE films, the migration rate into 3% acetic acid is significantly higher than into 10% ethanol (P < 0.05). However, the opposite trend is observed for the CPP film. The migration rates through the four samples into 10% ethanol and 3% acetic acid range from 13.26% to 22.63% and from 10.53% to 28.9%, respectively (Fig. 3). 3.2.4. Inner-layer dependence of the toluene migration rate The results obtained here indicate that irrespective of the simulant and of the composition of the films, increasing their thickness improves their barrier performance and better impedes the migration of toluene (Fig. 3). The migration rates of toluene into isooctane, 50% ethanol, and 3% acetic acid are significantly higher through LDPE30 than through LDPE50 (P < 0.05). With 10% ethanol as the simulant however, the migration rates through LDPE30 and LDPE50 are not significantly different (P > 0.05). The migration rates into all four simulants are higher through CPP30 than through CPP50 (P < 0.05). Furthermore, as attested by the migration rates, CPP films form a better barrier in this context than LDPE ones do. Indeed, at 30  C, the 30 mm thick CPP film still forms a more effective barrier than does the 50 mm thick LDPE film. Other than for

Fig. 1. Toluene migration rate as a function of temperature through LDPE30 (-), LDPE50 (B), CPP30 (:), and CPP50 (A) films into (A) isooctane, (B) 50% ethanol, (C) 10% ethanol, and (D) 3% acetic acid. For each film, migration rates that are significantly different are indicated with a different letter, lower case for P < 0.05 and upper case for P < 0.01.

168

N. Chang et al. / Food Control 59 (2016) 164e171

Fig. 2. Toluene migration rate at 40  C as a function of time through LDPE30 (-), LDPE50 (B), CPP30 (:), and CPP50 (A) films into (A) isooctane, (B) 50% ethanol, (C) 10% ethanol, and (D) 3% acetic acid.

Table 4 Time taken for toluene migration to reach equilibrium at 40  C and the corresponding migration rate in different samples. Sample

LDPE30 LDPE50 CPP30 CPP50

(Bhunia, Sablani, Tang, & Rasco, 2013). It reflects the distribution of the migrating matter between the two phases. The formula is as follows:

Equilibration time (h)/Migration rate(%) Isooctane

50% ethanol

10% ethanol

3% acetic acid

48/76.27 96/59.50 72/51.26 144/49.02

72/63.61 120/45.06 96/31.03 168/27.17

96/20.90 144/20.01 144/17.38 192/13.48

72/28.07 144/23.56 120/14.47 168/10.01

the 10% ethanol simulant, the migration rates through the former are always higher than through the latter. In terms of toluene migration rates therefore, the different samples studied here are ranked as follows: LDPE30 > LDPE50 > CPP30 > CPP50. 3.2.5. Toluene concentration dependence of the toluene migration rate The partition coefficient, KP,F is defined as the ratio of the concentration (CP,e) of a migrating compound in a polymer to its concentration (CF,e) in the food or food stimulant at equilibrium

Fig. 3. Toluene migration rates through different films into different simulants after 10 d at 40  C. Upper- and lower-case letters to indicate significant differences (P < 0.05) in the toluene migration rates measured for the same film in different simulants, and for the same simulant through different films, respectively.

KP;F ¼

CP;e : CF;e

Table 5 shows that the LDPE30 with initial toluene contents B (approx. 2.6 mg/m2) and C (approx. 3.6 mg/m2), the KP,Fs between LDPE30 into isooctane are 0.3284 and 0.3036 respectively, and though LDPE50 into 50% ethanol, 1.321 and 1.091 respectively. These differences are the only ones of note in the results obtained for samples with different initial toluene contents. There is no significant difference between the partition coefficients measured for CPP30 and CPP50 with different initial toluene concentrations (P < 0.05). This indicates that within the range studied here, the initial toluene content of the laminated films has almost no influence on its subsequent migration rate. 3.2.6. Infrared spectrum analysis Fig. 4 shows the infrared spectra obtained for the LDPE film. In Fig. 4A, strong absorption peaks characteristic of polyethylene are observed at 2937.39 cm1, 1463.61 cm1, and 720.23 cm1, with the peak at 2937.39 cm1 corresponding to CeH stretching vibrations. This peak is particularly intense as a result of the high concentration of CeH groups in the polyethylene film. The peak at 1463.61 cm1 is assigned to the deformation vibrations of methylene groups (CH2), while their rocking vibrations lead to the peak at 720.23 cm1 (Liu & Liu, 2008; Willeams & Fleing, 2001). Fig. 4B shows the infrared spectrum obtained from the same sample but treated with isooctane. The absorption peaks shift to lower wavenumbers, the one at 2937 cm1 in particular. A shift to lower wavenumbers/frequencies highlights a reduction in the absorption power and intermolecular forces. This suggests that isooctane weakens the intermolecular forces in the plastic films. Fig. 5 shows the infrared spectra obtained for the CPP film. The peaks in Fig. 5A at 2925.48 cm1 and 2838.31 cm1 arise from the stretching vibrations of methyl (CH3) and CeH groups, respectively. The peaks at 1457.31 cm1 and 1377.00 cm1 are assigned to CeH

N. Chang et al. / Food Control 59 (2016) 164e171

169

Table 5 The partition coefficients for toluene between polymer and food simulant. Sample

Toluene initial contenta

K value ± SDb Isooctane

LDPE30 LDPE30 LDPE30 LDPE50 LDPE50 LDPE50 CPP30 CPP30 CPP30 CPP50 CPP50 CPP50 a b

A B C A B C A B C A B C

0.3111 0.3284 0.3036 0.681 0.676 0.702 0.951 0.820 0.922 1.040 0.997 1.213

± ± ± ± ± ± ± ± ± ± ± ±

50% ethanol 0.008ab 0.011b 0.006a 0.009a 0.021a 0.012a 0.011a 0.038a 0.051a 0.087a 0.072a 0.079a

0.572 0.569 0.581 1.219 1.321 1.091 2.222 2.099 2.228 2.681 2.538 2.762

± ± ± ± ± ± ± ± ± ± ± ±

0.010a 0.017a 0.008a 0.012ab 0.057b 0.032a 0.038a 0.070a 0.063a 0.043a 0.049a 0.076a

10% ethanol 3.785 3.879 3.716 3.998 4.061 4.103 4.754 4.579 4.865 8.418 8.706 8.891

± ± ± ± ± ± ± ± ± ± ± ±

0.162a 0.158a 0.109a 0.071a 0.168a 0.103a 0.105a 0.132a 0.125a 0.289a 0.231a 0.207a

3% acetic acid 2.563 2.482 2.529 3.244 3.492 3.381 5.911 6.179 5.882 8.990 9.046 8.791

± ± ± ± ± ± ± ± ± ± ± ±

0.093a 0.109a 0.083a 0.112a 0.096a 0.126a 0.137a 0.297a 0.261a 0.290a 0.305a 0.226a

See Table 4 for the specific values. SD is the standard deviation of triplicate measurements; different letters within a column indicate results that are significantly different (P < 0.05).

bending vibrations. The characteristic [CH2CH(CH3)]n peaks due to out-of-plane CeH bending appear at 997.70 cm1. Three bands related to crystallization are observed at 1303.13 cm1, 1167.13 cm1, and 973.39 cm1. As shown in Fig. 5B, treating the CPP film with isooctane does not lead to changes in the resulting infrared spectrum, other than the absorption peak at 2925 cm1 moving to lower wavenumbers. This indicates that the molecular structure of the CPP film changes little on exposure to isooctane. Furthermore, the crystalline bands illustrate the better crystallinity and therefore the better barrier properties of the CPP film, compared with those of the LDPE film. 4. Discussion These tests reveal that temperature is a key accelerating factor for toluene migration, with higher temperature invariably leading to higher toluene migration rates (Fig. 1). In a previous study, atocopherol was found to migrate more rapidly into milk powder through multilayer films at higher temperatures (Granda-Restrepo et al., 2009). Similarly, the plasticizer content of sunflower-seed oil,

coming from the polyvinyl chloride packaging, was significantly higher at 40  C than at 20  C (Zhu & Wang, 2006). According to free volume theory, two criteria define govern whether migrating molecules are mobile in polymers: first, the polymers need to have sufficient free volume for the migrating molecules to circulate; second, in order to enter the free volume, the migrating molecules need to have enough energy to overcome the attraction from surrounding molecules (Cheng, 2011). Thereby, increasing the temperature increases the activity of polymer chains, which allows more toluene molecules to be accommodated in the vacated space. The energy of the toluene molecules also increases with the temperature. These two factors explain the increase in the migration rate observed here (and elsewhere) at higher temperatures. Our results also suggest that different food simulants have different effects on toluene migration, with higher rates observed for isooctane and 50% ethanol than for 10% ethanol and 3% acetic acid (Fig. 3). This may be because of the swelling of the packaging materials caused by the reverse permeation of the foodstuffs into the films. Swelling generally occurs when small migrating molecules embed themselves between the larger ones, weakening the

Fig. 4. Infrared spectra of LDPE films (A) as prepared, and (B) after treatment at 40  C for 10 d with isooctane.

170

N. Chang et al. / Food Control 59 (2016) 164e171

Fig. 5. Infrared spectra of CPP films (A) as prepared, and (B) after treatment at 40  C for 10 d with isooctane.

interactions between the latter and leading to volume expansion (Grassi, Colombo, & Lapasin, 2001; Hedenqvist & Gedde, 1999). In this study, all four simulants caused swelling, with isooctane having the most severe effect. Swelling alters the structure and morphology of the polymer network, increasing the activity of the molecular chains and the internal free volume. This explains the higher migration rates measured here into isooctane. Concerning the difference between the results obtained with 3% acetic acid and 10% ethanol, we postulate that the greater swelling effect of the former on LDPE films accelerates toluene migration. Similarly, the migration rates measured with 10% ethanol as the simulant differ less because of its lower swelling effect on LDPE and CPP films. This study also suggests that simulant polarity is important in determining the toluene migration rate. As captured in the popular aphorism “like dissolves like,” solutes are expected to be more soluble in solvents with a similar polarity. Toluene is weakly polar, and since isooctane is the only non-polar simulant among the four investigated here, toluene should dissolve more easily in isooctane and therefore migrate at higher rates into it. Similarly, the enhanced migration described previously of caprolactam through multilayer polyamide films into 3% acetic acid may be due to the lix, Manzoli, similar polarity of caprolactam and acetic acid (Fe Padula, & Monteiro, 2014). The tests performed here show furthermore that the migration rate of toluene depends on the material used for the inner layer of the laminated films, with CPP forming a more effective barrier than LDPE (Fig. 3). This is in agreement with Huang, Wang, Hu, Zhu, and Wang (2013), who found that the barrier properties of PP (polypropylene) were superior to those of LDPE. In this study, PP and LDPE are both used to coat packaging paper but these results suggest that the migration of substances is not the same through the two polymers (Huang et al., 2013). Both CPP and LDPE are polymers with high molecular weights within which crystalline and amorphous regions co-exist. Crystalline regions form complete arrays of lattice structure with strong interactions between the molecular chains. In contrast, the disordered arrays of molecular chains in amorphous regions are primarily arranged in irregular coils with only weak interactions between the chains (Zhang, Li, An, & Jiang, 2013). At room temperature, in terms of the phase diagram,

the amorphous polymer regions are close to the glasserubber transition. The activity of the chains is relatively high, allowing toluene molecules to migrate through the free volume in the polymer chains. Our results obtained here demonstrate that toluene migration is much slower through CPP than through LDPE films. In terms of the molecular structure, a possible explanation is that the CPP films are more crystalline (less amorphous) than the LDPE films, with stronger interactions between the molecular chains that better impede toluene migration. It is also clear from these results that thicker films are more effective at preventing toluene migration (Fig. 3). Although thicker films afford more free volume to the toluene molecules, the migration paths become less direct in the thickness direction and it takes more time for the molecules to pass through the film, thereby lowering the migration rates (Huang & Wang, 2009). In the experiments conducted for this study, the toluene migration rates increased over time before plateauing (Fig. 2). Similar results have been described previously in the literature, for the migration of silver from nanosilver-plastic packaging for instance (Tian et al., 2014). Elsewhere, the migration at 20  C of atocopherol through LDPE films into corn oil was found to plateau after one week (Graciano-Verdugo et al., 2010). A possible explanation for these results is that once the solubility limit of toluene is reached, it stops migrating, leading to an equilibrium state. However, the present study also demonstrates that the initial toluene content in the laminated films does not significantly affect the partition coefficient (K) for toluene. This indicates that the simulants investigated here are not saturated with toluene within the present concentration range. Further investigations are therefore required to interpret these results, to understand how the molecular migration dynamics lead to this plateauing. For instance, the solubility of migrating residues at packaging/simulant interfaces is an interesting subject for future studies. 5. Conclusions This study investigated the migration of toluene through laminated films into food simulants, concentrating on the effects on the migration rate of different temperatures, exposure times, inner

N. Chang et al. / Food Control 59 (2016) 164e171

layer materials, simulants, and initial toluene contents. For all four simulants, the toluene migration rate increased with the temperature, nonetheless reaching a plateau at between 48 h and 192 h of exposure depending on the simulant/film combination. Increasing the thickness of the films lowered the migration rate of toluene into the simulants. The CPP films formed more effective toluene barriers than the LDPE ones. Isooctane was the simulant for which the highest migration rates were measured and in this regard, in terms of the migration rates through LDPE, the four simulants investigated here are classed as follows: isooctane >50% ethanol >3% acetic acid >10% ethanol. Through CPP however, the corresponding order is: isooctane >50% ethanol >10% ethanol >3% acetic acid. Within the range of concentrations tested here, increasing the initial toluene content of the laminated films had almost no measurable effect on the subsequent migration rates. Infrared spectra recorded on LDPE and CPP films exposed and those not exposed to isooctane suggest that the somewhat lower intermolecular forces in LDPE due to the lower crystallinity of LPDE make these films more susceptible to the effect of food simulants than CPP films are. To avoid the migration of solvent residues into food, this study shows that a CPP inner film is preferable because it forms a more effective barrier. Furthermore, the data presented here will allow the relevant regulatory bodies to predict the toluene contents of foodstuffs, from which a risk assessment and the appropriate standards can be derived to ensure food safety and protect public health. Acknowledgments This work was supported by the National Natural Science Foundation of China (31370685). References €derhjelm, L. (2001). Migration of alkylbenzenes from Aurela, B., Ohra-aho, T., & So packaging into food an tenax. Packaging Technology and Science, 14, 71e77. Bhunia, K., Sablani, S. S., Tang, J., & Rasco, B. (2013). Migration of chemical compounds from packaging polymers during microwave, conventional heat treatment, and storage. Comprehensive Reviews in Food Science and Food Safety, 12, 523e545. Bowen, S. E., & Hannigan, J. H. (2013). Binge toluene exposure in pregnancy and preweaning developmental consequences in rats. Neurotoxicology and Teratology, 38, 29e35. Cheng, F. L. (2011). Research of mathematical model for plastic packaging materials migration. Journal of Beijing Technology and Business University (Natural Science Edition), 29, 61e63. ^ F., Araújo, V. A., & Rodrigues, R. (2014). Coltro, L., Pitta, J. B., Costa, P. A., Perez, M.A. Migration of conventional and new plasticizers from PVC films into food simulants: a comparative study. Food Control, 44, 118e129. Commission regulation 10/2011/EU of 14 January 2011 on plastic materials and articles intended to come into contact with food. Official Journal of the European Union L Series, 12, 1e89. Council Directive 82/711/EEC of 18 October 1982 laying down the basic rules necessary for testing migration of the constituents of plastic materials and articles intended to come into contact with foodstuffs. Official Journal of the European Union L Series, 297, 26e30. Council Directive 85/572/EEC of 19 December 1985 on laying down the list of simulants to be used for testing migration of constituents of plastic materials and articles intended to come into contact with foodstuffs. Official Journal of the European Union L Series, 372, 14e21. Dong, W. L., Li, Y. J., & Zhang, W. F. (2011). Migration analysis of residual solvents of food packaging material in food. Packaging Engineering, 32, 27e29. European Committee for Standardization (CEN). (2004). EN 13130e1: Materials and articles in contact with foodstuffs d Plastics substances subject to limitation. In Patr1: Guide to the test methods for the specific migration of substances from plastics to food and food simulants and the determination of substances in plastic and the selection of conditions of exposure to food simulants. Brussels (p. 65). lix, J. S., Manzoli, J. E., Padula, M., & Monteiro, M. (2014). Evaluation of Fe different conditions of contact for caprolactam migration from multilayer

171

polyamide films into food simulants. Packaging Technology and Science, 27, 457e466. Ferrara, G., Bertolodo, M., Scoponi, M., & Ciardelli, F. (2001). Diffusion coefficient and activation energy of Irganox 1010 inpoly (propylene-co-ethylene) copolymers. Polym Degrad Sta, 73, 411e416. General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China. (2006). Review guidelines for the manufacturing license of food-grade plastic package, container, tools and other products. General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China. (2008). GB/T 10004-2008 plastic laminated films and pouches for packaging d dry lamination and extrusion lamination. General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China. (2009). GB/T 23296.1e2009 materials and articles in contact with foodstuffs-plastics substances subject to limitation-guide to test methods for the specific migration of substances from plastics to foods and food simulants and the determination of substances in plastics and the selection of conditions of exposure to food simulants. Goulas, A. E. (2001). Overall migration from commercial coextruded food packaging multilayer films and plastics containers into official EU food simulants. European Food Research and Technology, 212, 597e602. rate, P., IslasGraciano-Verdugo, A. Z., Soto-Valdez, H., Peralta, E., Cruz-Za Rubio, A. R., S anchez-Valdes, S., et al. (2010). Migration of a-tocopherol from LDPE films to corn oil and its effect on the oxidative stability. Food Research Inernational, 43, 1073e1078. Granda-Restrepo, D. M., Soto-Valdez, H., Peralta, E., Troncoso-Rojas, R., Vallejrdoba, B., Ga mez-Meza, N., et al. (2009). Migration of a-tocopherol from an Co active multilayer film into whole milk powder. Food Research International, 42, 1396e1402. Grassi, M., Colombo, I., & Lapasin, R. (2001). Experimental determination of the theophylline diffusion coefficient in swollen sodium-alginate mebranes. Journal of Controlled Release, 76, 93e105. Hedenqvist, M. S., & Gedde, U. W. (1999). Parameters affecting the determination of transport kinetics data in highly swelling polymers above Tg. Journal of Polymer, 40, 2381e2393. Huang, X. L., & Wang, Z. W. (2009). The migration of photoinitiator in UV ink of paper coating with PE into food simulants. Polymer Materials Science and Engineering, 25, 97e100. Huang, X. L., Wang, Z. W., Hu, C. Y., Zhu, Y., & Wang, J. (2013). Factors affecting migration of contaminants from paper through polymer coating into food simulants. Packaging Technology and Science, 26, 23e31. Katan, L. L. (1996). Migration from the food contact materials (pp. 77e110). London: Springer - Verlag New York Inc.. Liu, R. Z., & Liu, J. S. (2008). .Identification of plastic composition by IR spectra. Plastics and Technology, 36, 72e77. Mao, X. Q., Zheng, S. L., Yu, L. J., Jiang, Z. B., & Chen, J. P. (2008). Comparative study of management system of plastic food-packaging materials worldwide. Packaging Engineering, 29, 42e52. Meng, Z., Liao, X., Sun, D. D., Liao, Y. P., & Liu, H. W. (2007). Migration of alkylbenzene component from ink used in food packaging material by GC-MS. Chemical Journal of Chinese Universities, 28, 1039e1042. Regulation 1935/2004/EC of the European parliament and of the council of 27 October 2004. Official journal of the European Union C Series, 338, 4e17. Reinas, I., Oliveira, J., Pereira, J., Machado, F., & Pocas, M. F. (2012). Migration of two antioxidants from packaging into a solid food and into Tenax. Food Control, 28, 333e337. Tian, H. J., Lin, Q. B., Guo, J., Li, B., Wang, R. Z., Liu, G. H., et al. (2014). Impact of additives on migration of silver from nanosilver-plastic food packaging to food simulants. Food Science, 5, 8e12. Torres, A., Guarda, A., Moraga, N., Romero, J., & Galotto, M. J. (2012). Experimental and theoretical study of thermodynamics and transport properties of multilayer polymeric food packaging. European Food Research and Technology, 234, 713e722. Twede, D., & Goddard, R. (1998). Introduction to plastics,polyolefins-polyethylene and polypropylene, styrenic plastics (pp. 69e100). UK: Pira International. Vera, P., Canellas, E., & Nerín, C. (2014). Migration of odorous compounds from adhesives used in market samples of food packaging materials by chromatography olfactometry and mass spectrometry (GC-O-MS). Food Chemistry, 145, 237e244. Willeams, D. H., & Fleing, I. (2001). Spectroscopic methods in organic chemistry (5th ed.). New York: McGraw-Hill (Chapter 2). Xu, C. X., Yang, Y., Gao, J. W., & Lu, J. (2008). Simultaneous determination of 15 kinds of residual organic solvents in laminated films used for food packages by headspace gas chromatography. Food Science, 29, 496e498. Zhang, Y., Li, H. F., An, L. J., & Jiang, S. C. (2013). Progress in studies on structure and relaxation behavior of the amorphous phases in crystalline polymers. Acta Polymerica Sinica, 4, 462e471. Zhu, Y., & Wang, Z. W. (2006). Migration of plasticizers in the PVC film applied in food packaging. Packaging Engineering, 27, 40e41.