Influence of some packaging materials and of natural tocopherols on the sensory properties of breakfast cereals

Abstract

The combined effect of natural antioxidants and packaging materials on the quality decay of breakfast cereals during storage was evaluated. Corn flakes were produced on industrial scale, using different packages and adding natural tocopherols to the ingredients, and stored for 1 year. The samples were then submitted to sensory analysis and HS-solid phase microextraction/gas chromatography/mass spectrometry (SPME/GC/MS) analysis. The packaging had a significant influence on the sensory profile of the aged product: metallized polypropylene gave the highest levels of oxidation compounds and sensory defects. The sensory profile was improved using polypropylene and especially high-density polyethylene. Natural tocopherols reduced the sensory decay of the flakes and the oxidative evolution of the volatile profile. They gave the most remarkable improvement in polypropylene (either metallized or not) packs. Polypropylene showed a barrier effect on the scalping of volatiles outside of the pack. This led to higher levels of oxidation volatiles and faster rates of the further oxidative processes involving the volatiles.

Keywords

Breakfast cereals natural antioxidants food quality packaging/antioxidants interaction

Introduction

The quality parameters of extruded products such as breakfast cereals are a function of the processing conditions they are submitted to (Doğan and Karwe, 2003; Hernández-Díaz et al., 2007; Ilo and Berghofer, 1999; Sumithra and Bhattacharya, 2008). When they undergo aging, both physical (Gonzalez et al., 2010) and chemical (Paradiso et al., 2008) processes take place. In both cases, sensory properties are directly affected, with a subsequent quality decay.

From a chemical point of view, the low moisture content, high surface area and potential for metal contamination as a result of extrusion increase the susceptibility to oxidation of extruded products, though their lipid content usually do not exceed 6–7%. The effectiveness of natural antioxidants in slowing down oxidative processes, preventing therefore the onset of off-flavours during storage has been studied, particularly in model systems (Butt et al., 2003; Camire et al., 2005, 2007; Viscidi et al., 2004; White et al., 2010). In our previous works (Paradiso et al., 2008, 2009) the use of different natural antioxidants was evaluated in the production of industrial-scale corn flakes. Tocopherols resulted to be the most effective during long-term (1 year) storage of the packed product.

On the other hand, when considering real systems, as a packed food, interactions of antioxidants with other variables, as packaging, are to be considered.

Active packaging is an example of this interaction. A wide literature exists about antioxidant-releasing packaging materials (Lopez-de-Dicastillo et al., 2012; Nerín, 2010; Nerín et al., 2008). As regards the specific use of antioxidant-releasing packaging for breakfast cereals, Laermer et al. (1993) showed that the addition of α-tocopherol to high-density polyethylene (HDPE)-ethylene/vinyl acetate (EVA)-HDPE flexible packaging system reduced the “plastic” taste and preserved the fresh taste of breakfast cereals. A direct effect in inhibiting the oxidation of the fatty fraction should be exerted by a release of the antioxidant to the fatty fraction of the food. Wessling et al. (1998) studied the migration/sorption behaviour of butylated hydroxytoluene (BHT) and α-tocopherol in packaging material in contact with fatty food simulants. They found that α-tocopherol, being a bigger molecule than BHT, showed lower mobility. Yet, the concerns regarding the use of synthetic antioxidants have moved the interest towards the use of natural additives (such as α-tocopherol). Another study by Wessling et al. (1999) pointed out that these interactions are influenced by both the nature of the food and the packaging material: the migration of α-tocopherol from plastic materials is generally much higher when in contact with fatty food products than with water-based foods; also, unlike low-density polyethylene (LDPE), polypropylene (PP) did not show any interactive tendencies towards any of the foods, and the α-tocopherol content in the PP film remained more or less unaffected. Being commercial breakfast cereals such as corn flakes, a dry food obtained from a water-based mixture and containing very low amounts of fat, a controlled release of α-tocopherol appears quite difficult. Therefore, it should be preferable adding antioxidants to the dough and subsequently packing the product.

Butt et al. (2003) evaluated the addition of butylated hydroxyanisole (BHA) and BHT to breakfast cereals with nuts, packed with bi-oriented polypropylene (BOPP), HDPE or aluminium foil. Flakes were stored up to 6 months and submitted to sensory evaluation: samples in aluminium foil with BHT were found best.

On the other hand, information about the combined effect of packaging with natural antioxidants added to breakfast cereals is still lacking.

The present work was aimed to assess the combined effect of natural antioxidants and packaging materials on the sensory decay, due to the lipid oxidative degradation, involving breakfast cereals during storage. In particular, three different packages and natural tocopherols were evaluated in a long-term-storage experiment carried out on corn flakes produced on industrial scale.

Materials and methods

Sample preparation

Corn flakes were produced using an industrial production line (Figure 1). Raw materials (corn flour, sugar, salt, malt, water) were dosed in mixing machinery in such quantities as to obtain a mixture of 400 kg. After steam pre-cooking in the kneader, the mixture was extruded-cooked in a twin-screw extruder. The pellets obtained were directed to a sieve, and then flaked. After a second passage on a sieve, the flakes were toasted and finally cooled. The fat content of the flakes obtained was about 2.5%.

The flakes produced through the above technology (control, C) were compared with flakes (T) added with 187.5 mg/kg of natural tocopherols. This was the best solution selected in a previous research, aimed to compare different natural antioxidants (Paradiso et al., 2009). A commercial formulate containing 30% of natural tocopherols (14% D-αtocopherol; 2% D-γ-tocopherol; 60% D-γ-tocopherol; 24% D-γ-tocopherol) sprayed onto gum acacia (Covi-ox® T-30P) was purchased from Cognis Corporation (Cincinnati, OH, USA).

The two typologies of flakes were packed in three different packages:

coextruded polypropylene/polypropylene cast coupled film (CartonPack S.r.l., Bari, Italy) – with permeability to O₂ < 1100 cm³ m⁻² 24 h⁻¹ (25 ℃, 0% RH) and permeability to water <5 g m⁻² 24 h⁻¹ (38 ℃, 90% RH) – PP;

coextruded HDPE with a ionomer layer – with permeability to O₂ < 110 cm³ m⁻² 24 h⁻¹ (23 ℃, 0% RH) and permeability to water <13 g m⁻² 24 h⁻¹ (38 ℃, 90% RH) – HDPE, coupled with a cardboard box;

coextruded polypropylene/coextruded metallized polypropylene coupled film (CartonPack S.r.l., Bari, Italy) – with permeability to O₂ < 80 cm³ m⁻² 24 h⁻¹ (25 ℃, 0% RH) and permeability to water <1 g m⁻² 24 h⁻¹ (38 ℃, 90% RH) – PPmet.

The typologies of packaging PP and HDPE were those ordinarily used for this kind of product (the former is usually found in discounts, while the latter is the most commonly adopted). The PPmet film, instead, is usually adopted for other typologies of product (e.g. potato chips): it was considered being the one with the lowest values of gas permeability.

On the whole, six typologies of samples were obtained by the combination of the two variables considered: C-PP, C-HDPE, C-PPmet, T-PP, T-HDPE, T-PPmet.

Four packs of C-PP, one of C-HDPE, one of C-PPmet, two of T-PP, one of T-HDPE and one of T-PPmet were sampled and stored at room temperature and under normal conditions of light for 1 year. In particular, the room average temperature was in the range 20–24 ℃ while the relative humidity was in the range 40–70% throughout the storage test. The light exposure simulated the conditions of a supermarket shelf and were the following: artificial light; wavelength range of neon lamps; lighting duration of 12 h. A different number of packs was sampled since particular attention was paid to PP, which showed the highest permeability to oxygen.

After the storage period samples were analyzed by sensory evaluation and solid phase microextraction-gas chromatography/mass spectrometry (SPME-GC/MS) analysis of the volatile oxidation compounds.

Sensory evaluation

The sensory evaluation was carried out as reported in our previous work (Paradiso et al., 2009). International Organization for Standardization (International Organisation for Standardisation, 1999) regulations were used to identify the steps required to implement the sensory analysis of a product. They included (1) the identification of the descriptors qualifying flavours and off-flavours that appear in corn flakes during storage, considering the descriptors that are indices of the lipid oxidation. The descriptors adopted were the following: fresh product odour (FPO), rancid odour (RO), pungent odour (PO), stale odour (SO), fresh product flavour (FPF), rancid flavour (RF), overall acceptability (OA); (2) the arrangement of a form for sensory analysis. The intensity of every attribute was expressed on a 10-cm unstructured linear scale. The left end of the scale corresponded to the least intensity (value 0) of the attribute, while the right end corresponded to the maximum intensity (value 10); (3) the standardization of a method for the preparation and tasting of samples. The tasting method was standardized and subdivided into several steps: at the beginning of the panel session, each panellist was asked to evaluate the odour descriptors, and then he/she was asked to assess the taste descriptors. Samples were given, after the opening of the pack, in 200 mL PE glasses covered with aluminium film; (4) formation and selection of the panel. The panel group was selected among 15 researchers and technicians of the research laboratory, undergraduate and PhD students, habitual consumers of corn flakes, after a cycle of lectures on the basic notions of sensory analysis and some tests to monitor their performance in recognizing the off-flavour of corn flakes. Six panellists were selected (three male and three female members; age ranging from 25 to 44). Panellists were submitted to training in several preliminary sessions, also by evaluating fresh and aged flakes, in order to familiarize with flavours and with the vocabulary and the intensity range of the selected sensory attributes; (5) execution procedure of the panel test. The samples were randomly coded using a two-digit number. Each sample was submitted at room temperature and anonymously to all panel members. Panellists received a tray containing samples, a glass of water and an evaluation form with a pencil. One sample of each typology was scored by panellists.

Headspace analysis

Volatiles were extracted by solid phase micro-extraction (SPME) as reported in Paradiso et al. (2008). In particular, samples were weighed (10 ± 0.05 g) in a 50-mL vial, closed by butylic rubber septa and an aluminium seal. Before extraction, stabilization of the headspace in the vial was obtained by equilibration for 30 min at 40 ℃. The extraction was carried out using a divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS) 50/30 µm SPME fiber assembly (Supelco, Bellefonte, PA, USA) at 40 ℃ for 30 min. The fiber was desorbed for 6 min in the injection port of the gas-chromatograph, operating in splitless mode. For the SPME analyses an Agilent 6850 gas-chromatograph equipped with an Agilent 5975 mass-spectrometer was used. Compounds were resolved on a Supelco capillary column SPB-624 (30 m × 0.25 mm × 1.4 µm), under the following conditions: injection port temperature, 250 ℃; helium pressure, 30 kPa; oven temperatures, 40 ℃ for 2 min then 5 ℃ min⁻¹ to 230 ℃ and final isothermal for 10 min.

Mass detector was set at the following conditions: detector voltage, 500 V; interface temperature 250 ℃; source temperature 250 ℃; ionization energy 70 eV; emission 200 Å, scan range 30–270 amu. Peak identification was performed comparing retention times with those of internal databases, by Linear Retention Index (LRI) and by computer matching with the reference mass spectra of NIST and Wiley libraries. In some cases, identification was performed only by matching with the reference mass spectra. To compensate for the variation in the performance of the SPME fiber and the MS-detector during the 12-month storage period, all detector responses of duplicated samples were standardized to the response of an external standard. Semi-quantitative data (electronic units of the integrated peak area) were used.

Statistical analysis

Analysis of variance (ANOVA), followed by Tukey Test for multiple comparisons, Principal Components Analysis (PCA) and Partial Least Squares Regression (PLSR) were carried out on the experimental data by the XLStat software (Addinsoft SARL, New York, NY).

RESULTS AND DISCUSSION

Figure 2 reports the characterization of the sensory profiles of the 1-year-old flakes, obtained by means of the two-way ANOVA models (product and judge variables). If the coefficient of a descriptor results to be significantly different from zero for a typology of flakes (as can be seen from Table 1, reporting the p-values for all the model coefficients), this means that the sensory profile of that typology of flakes is significantly different from the average sensory profile of the flakes. The bar charts in the Figure report the ANOVA coefficients for each typology of flakes. Coloured bars are related to positive and negative coefficients significantly differing from zero (p < 0.05).

Table 1.

Significance (p-values) of the model coefficients for the sensory product characterization.^a

C-PP	C-PPmet	C-HDPE	T-PP	T-PPmet	T-HDPE
FPO	0.035	0.310	0.906	0.051	0.146	0.066
PO	0.087	0.065	0.146	0.237	0.624	0.278
RO	0.052	0.020	0.080	0.245	0.189	0.073
SO	0.133	0.145	0.418	0.635	0.404	0.253
FPF	0.010	0.109	0.260	0.014	0.019	0.030
RF	0.049	0.026	0.031	0.159	0.061	0.061
SF	0.044	0.089	0.269	0.113	0.856	0.466
OA	0.216	0.308	0.612	0.434	0.802	0.886

FPO: fresh product odour; PO: pungent odour; RO: rancid odour; SO: stale odour; FPF: fresh product flavour; RF: rancid flavour; SF: stale flavour; OA: overall acceptability. C-PP, control flakes packed in polypropylene; T-PP, flakes with tocopherols packed in polypropylene; C-PPmet, control flakes packed in metallized polypropylene; T-PPmet, flakes with tocopherols packed in metallized polypropylene; C-HDPE, control flakes packed in high-density polyethylene; T-HDPE, flakes with tocopherols packed in high-density polyethylene.

If a p-value is bold (p < 0.05) this means that the sensory profile of that typology of flakes is significantly different from the average sensory profile of the flakes for the corresponding descriptor.

As can be seen in the figure, the C-PP flakes were characterized, with respect to the mean profile of the flakes, by more intense perceptions of PO and RF and, consequently, by a lower OA. The defect descriptors PO and RF were previously reported as characterizing corn flakes in the most advanced phases of ageing (Paradiso et al., 2008). Packing the control flakes in HDPE, the resulting sensory profile was sensibly improved, with a lower perception of the PO and an increase in the OA. The worst sensory properties were detected by the assessors in the C-PPmet flakes, which showed the lowest FPO and OA scores, while PO, RO and SO, as well as the RF, resulted to characterize significantly their profile.

The addition of tocopherols improved the sensory profile of the flakes in all the typologies of packaging. Flakes packed in PP were no more characterized by PO and RF; also the OA was no more lower than in the average sensory profile. The sensory profile of T-HDPE flakes too was improved by tocopherols, since the OA increased and the coefficient for the RO was significantly negative. Finally, all the defects found in the control flakes stored in PPmet, as well as the lack of freshness and acceptability, did not characterize any more the product added with tocopherols and packed with the same film. Principal components analysis was performed on the data of the sensory analysis and gave the results reported in the loading plot (a) and the score plot (b) of Figure 3. The PC1 could explain a great part of variability (over 90%). All the descriptors, in fact, were strictly correlated: in particular, all the defect descriptors were positively correlated with the PC1, being the flakes in an advanced stage of their shelf-life, characterized by the perception of both early defects (such as RO) and late defects (e.g. PO and RF; Paradiso et al., 2008). As a consequence, the freshness descriptors and the acceptability were negatively correlated with the PC1. Along the PC1, the effects of the antioxidant and packaging variables can be observed. The flakes without tocopherols and stored in the PP films are placed on the extreme right, having the worst sensory properties (particularly the C-PPmet flakes). A second group of flakes, placed roughly near the origin of the PC1 axis, includes the flakes added with tocopherols and stored in both the PP films. The T-PP flakes, having slightly better sensory properties than T-PPmet, are placed near the third group, including the flakes packed in HDPE, both without and with tocopherols.

The results of the sensory analysis showed that a remarkable improvement of the sensory properties was obtained adding tocopherols to the corn flakes packed in PP and PPmet films. On the other hand, HDPE-packed flakes had in any case a better sensory profile, so the improvement due to the use of antioxidants was slight.

Table 2 reports the volatile compounds analyzed by the SPME-GC/MS technique and the results of the two-way ANOVA (antioxidant and packaging variables) performed on the semi-quantitative data. Twenty-four compounds were considered, deriving from the oxidation of lipids. Not all the models resulted to be significant, due to a certain variability of the results. Nevertheless, some remarkable significant effects were observed. The packaging variable affected the amounts of 15 compounds, belonging to all the considered groups (i.e. aldehydes, alcohols, ketones, acids, lactones, furans). In almost all these cases, the flakes packed in PPmet showed the highest amounts. HDPE-packed samples presented the lowest amounts, in some cases with significant differences also from PP samples.

Table 2.

Analyzed volatile compounds. Results of the two-way ANOVA performed on the semi-quantitative data of the SPME-GC/MS analysis. The Table reports p values, standardized coefficients and the results of the post-hoc Tukey HSD test.^a

	LRI	p value			C		T		PP		PPmet		HDPE
	LRI	Model	AO	Film	SC	Tukey	SC	Tukey	SC	Tukey	SC	Tukey	SC	Tukey
Aldehydes
Pentanal	734	0.159	0.497	0.117	−0.20	ns	0.20	ns	−0.60	ns	0.83	ns	−0.23	ns
Hexanal	840	0.368	0.117	0.923	−0.59	ns	0.59	ns	−0.10	ns	0.17	ns	−0.07	ns
2-(E)-hexenal	907	<0.001	0.024	<0.001	0.32	A	−0.32	B	−0.32	b	1.27	a	−0.95	b
Heptanal	939	0.007	0.549	0.004	0.10	ns	−0.10	ns	0.03	b	1.14	a	−1.17	c
2-(E)-heptenal	1005	<0.001	0.272	<0.001	0.12	ns	−0.12	ns	−0.53	b	1.34	a	−0.81	b
Octanal	1039	0.024	0.583	0.012	0.12	ns	−0.12	ns	−0.16	b	1.17	a	−1.01	b
2-(E)-octenal	1115	<0.001	0.073	<0.001	0.24	ns	−0.24	ns	−0.47	b	1.31	a	−0.84	b
Nonanal	1152	0.111	0.658	0.063	0.12	ns	−0.12	ns	−0.03	ns	0.99	ns	−0.96	ns
2-(E)-nonenal	1229	0.028	0.011	0.145	0.73	A	−0.73	B	0.01	ns	0.58	ns	−0.60	ns
2,4-(E,E)-nonadienal*	1287	0.002	0.668	<0.001	−0.06	ns	0.06	ns	0.03	b	1.19	a	−1.22	c
Alcohols
1-pentanol	822	0.277	0.330	0.216	0.32	ns	−0.32	ns	−0.22	ns	0.83	ns	−0.61	ns
1-hexanol	920	0.002	0.724	0.001	−0.05	ns	0.05	ns	−0.20	b	1.28	a	−1.08	c
1-heptanol	1014	<0.001	0.028	<0.001	0.23	A	−0.23	B	−0.47	b	1.34	a	−0.87	b
1-octen-3-ol*	1020	<0.001	0.266	<0.001	0.13	ns	−0.13	ns	−0.34	b	1.33	a	−0.99	b
1-octanol	1120	0.298	0.440	0.196	0.26	ns	−0.26	ns	−0.61	ns	0.71	ns	−0.10	ns
Ketones
2-heptanone	933	<0.001	0.249	<0.001	−0.05	ns	0.05	ns	−0.69	b	1.31	a	−0.61	b
3-octen-2-one	1090	<0.001	0.088	<0.001	0.12	ns	−0.12	ns	−0.52	b	1.36	a	−0.84	b
3,5-octadien-2-one*	1133	0.082	0.064	0.211	−0.56	ns	0.56	ns	−0.54	ns	0.26	ns	0.28	ns
2,3-octanedione*	1138	<0.001	0.073	<0.001	0.25	ns	−0.25	ns	−0.47	b	1.30	a	−0.83	b
Acids
Pentanoic acid	978	0.023	0.005	0.314	0.85	A	−0.85	B	−0.22	ns	0.44	ns	−0.22	ns
Hexanoic acid	1073	0.003	0.002	0.031	0.74	A	−0.74	B	0.44	a	0.13	ab	−0.57	b
Lactones
γ-hexalactone*	1175	0.012	0.052	0.009	0.42	ns	−0.42	ns	−0.44	b	1.14	a	−0.70	b
Furans
2-butylfuran*	909	0.054	0.019	0.351	−0.73	ns	0.73	ns	−0.18	ns	0.49	ns	−0.31	ns
2-pentylfuran	1003	0.001	0.946	<0.001	0.01	ns	−0.01	ns	−0.49	b	1.32	a	−0.83	b
Total	<0.001	0.009	<0.001	0.35	A	−0.35	B	−0.09	b	1.20	a	−1.11	c

C: control flakes; T: flakes added with tocopherols; PP: polypropylene; PPmet: metallized polypropylene; HDPE: high-density polyethylene; LRI: Linear Retention Index; AO: antioxidant; SC: standardized coefficient; *: tentatively identified by matching with reference mass spectra.

Capital letters are used for comparison of antioxidants; small letters are used for the comparison of packagings. Different letters mean a significant difference at p < 0.05; ns: not significant.

Only six compounds were affected by the antioxidant variable: in particular, the levels of the volatile acids were interested. The standardized coefficients denote as the influence of the antioxidant variable on the amounts of these compounds was higher than that of the packaging variable. This shows that the effects of packaging and antioxidants were not overlapping: the former affected the overall levels of volatiles, while the latter showed a selective activity regarding the products of advanced stages of oxidation.

The principal components analysis was carried out also on the data regarding the volatile compounds. Figure 4 reports the loading plot (a) and the score plot (b) obtained from the principal components analysis. The first two principal components explained over 83% of the total variability. PC1, in particular, accounted for 67% of the variability. Almost all the volatile compounds (included the volatile acids) were positively correlated with the PC1, and in many cases there were strict correlations. So, the typologies placed on the right side of the corresponding score plot were characterized by high relative amounts of volatile compounds. The PC2 resulted to be, instead, positively correlated with the aldehydes pentanal and hexanal, and negatively correlated with the corresponding acids – i.e. pentanoic and hexanoic acids, which are products of further oxidation of the respective aldehydes (Palamand and Dieckmann, 1974; Grosch, 1982), deriving from linoleic acid hydroperoxides – so that the items placed in the upper part of the plot showed slower rates of the oxidative turnover aldehyde/acid during the 1-year storage; on the contrary, the items standing in the lower part of the plot were characterized by faster oxidative processes and higher amounts of volatile acids. The score plot shows that the PC1 distinguished the typologies of flakes according to the packaging. In fact, the HDPE packs were those with the lowest amounts of volatile compounds. The PPmet flakes were on the opposite side of the plot, remarkably far from the other typologies of flakes, thus denoting high levels of volatile oxidation compounds. Almost 16% of the data variability was explained by the PC2, which was not connected to the overall amount of volatile compounds, but to the balances in the oxidative pairs aldehyde/acid. Along this component, the typologies of flakes were located according to the presence/absence of added antioxidants: in the lower part the control flakes packed in PP films revealed higher levels of the products of further oxidation, while up along the PC2, the flakes added with tocopherols and both the typologies packed in HDPE denoted lower oxidative rates. This means that, though both antioxidant and packaging variables affected the levels of oxidation volatile compounds, they exerted different actions. In fact, the different packaging films used influenced the relative amounts of almost all the oxidation volatile products and not their proportions, while the used antioxidants acted on the evolution of the volatile profile due to further oxidative phenomena. This effect was more remarkable as the levels of volatile compounds became higher: it was more marked in PPmet than in PP flakes and in PP than in HDPE flakes. As regards the effect of antioxidants, it clearly seems to regard the slowing down of the oxidative phenomena. Hexanoic acid has been proved to play a key role in off-flavours related to long-term storage of such products. Sjövall et al. (1997) did not include it among the volatile compounds identified in extruded oats in storage tests carried out for 18 weeks. Guth and Grosch (1993) noticed, instead, an increase of hexanoic acid during storage, extended till 1 year, of oat extruded products with the subsequent onset of off-flavour. Our previous studies also demonstrated its relationship with the sensory decay of long-term stored corn flakes (Paradiso et al., 2008, 2009). The effect of the films, instead, requires other explanations. The amounts of volatile compounds do not increase according to the increase of permeability to oxygen (the PPmet film, in fact, showed the highest barrier properties): the different behaviours should not be mainly attributed to the availability of oxygen due to permeation. Also scalping of the volatile compounds could be involved. Flavour scalping is a term used to indicate «food flavour sorption at the inner surface of polymer packaging, or migration of components through the film caused by sorption at the upper side, diffusion within the polymer, and desorption at the down side. This transfer can be oriented from the food product to the surrounding medium (“out migration”), corresponding to a loss of aroma quality, or in the opposite direction (“in migration”), causing a flavour adulteration of foods» (Debeaufort and Voilley, 1994). Flavour (and off-flavour) losses are then a possible effect of scalping (Dury-Brun et al., 2008; Sajilata et al., 2007). In particular, the different films allowed in different amounts the negative migration of volatile compounds to the pack. The metallized PP film, therefore, should oppose a great barrier to these out migration phenomena, leading to a concentration of the volatile compounds in the headspace, and consequently to both the increase in the sensory perception, and the acceleration of further oxidation of the volatile compounds. This could also explain why the effectiveness of tocopherols was more appreciable in this packaging.

These considerations are pointed out by the partial least squares regression (PLSR) performed on the analytical data (see the correlations map in the Figure 5). The regression models were built considering the sensory descriptors as a function of the volatile compounds relative amounts. The regression was performed with four components.

Being nearby the boundaries of the circle, all the descriptors are well represented by the PLSR models. Also, a great part of volatile compounds showed to contribute significantly to the PLSR models. As regards the typologies of flakes, C-PPmet could be clearly associated to high amounts of volatile compounds, and in particular a prevalence of volatile acids and alcohols, on their turn correlated to the defect descriptors. The T-PPmet flakes showed high amounts of volatile compounds, too, but they were less related to acids and alcohols, and consequently to sensory defects. The flakes packed in PP denoted the same trend due to the use of tocopherols, but lower amounts of volatile compounds and a consequent improvement of the sensory profile. The T-PP flakes were close to the best typologies of flakes, packed in HDPE, which were placed in the correlations map far from the sensory defects and the oxidation compounds, and close to the descriptors of freshness and OA.

Conclusion

The type of packaging had a significant influence on the sensory profile of aged corn flakes: the flakes stored in metallized PP were characterized by intense sensory defects and scarce OA; the use of PP slightly improved the sensory profile, while flakes stored in HDPE showed the best sensory profile.

The use of natural tocopherols improved the sensory profile of the flakes, whichever was the packaging material. Nevertheless, the most remarkable improvement was observed in the flakes stored in PP (either metallized or not).

Also, the headspace composition, with special reference to the lipid oxidation volatile compounds, proved to depend both on the type of packaging and on the use of antioxidants. Tocopherols, in fact, slowed down the oxidative evolution of the volatile profile. As regards the packaging, PP, and much more metallized PP showed a barrier effect on the scalping of volatiles outside of the pack. This lead to higher levels of oxidation volatiles and faster rates of the further oxidative processes involving the volatiles (e.g. the aldehyde/acid turnover). This could explain also the results obtained from the sensory analysis.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Butt

Ali

Pasha

Hashmi

Dogar

(2003) Effect of different antioxidants and packaging materials on the storage stability of breakfast cereals. Internet Journal of Food Safety 2: 1–5.

Camire

Dougherty

Briggs

(2005) Antioxidant-rich foods retard lipid oxidation in extruded corn. Cereal Chemistry 82: 666–670.

Camire

Dougherty

Briggs

(2007) Functionality of fruit powders in extruded corn breakfast cereals. Food Chemistry 101: 765–770.

Debeaufort

Voilley

(1994) Aroma compound and water vapor permeability of edible films and polymeric packagings. Journal of Agricultural and Food Chemistry 42: 2071–2075.

Doğan

Karwe

(2003) Physicochemical properties of quinoa extrudates. Food Science and Technology International 9: 101–114.

Dury-Brun

Hirata

Guillard

Ducruet

Chalier

Voilley

(2008) Ethyl hexanoate transfer in paper and plastic food packaging by sorption and permeation experiments. Journal of Food Engineering 89: 217–226.

Gonzalez

Khalef

Wright

Okos

Hamaker

Campanella

(2010) Physical aging of processed fragmented biopolymers. Journal of Food Engineering 100: 187–193.

Grosch

(1982) Lipid degradation products and flavours. In: Morton

MacLeod

(eds) Food Flavours. Amsterdam: Elsevier, pp. 325–398.

Guth

Grosch

(1993) Odorants of extrusion products of oat meal – changes during storage. Zeitschrift fur Lebensmittel-Untersuchung und-Forschung 196: 22–28.

10.

Hernández-Díaz

Quintero-Ramos

Barnard

Balandrán-Quintana

(2007) Functional properties of extrudates prepared with blends of wheat flour/pinto bean meal with added wheat bran. Food Science and Technology International 13: 301–308.

11.

Ilo

Berghofer

(1999) Kinetics of colour changes during extrusion cooking of maize grits. Journal of Food Engineering 39: 73–80.

12.

International Organisation for Standardisation (ISO, 1998). Sensory analysis. methodology. general guidance for establishing a sensory profile. ISO Standard no. 13299. Geneva, Switzerland: ISO.

13.

Laermer SF, Young SS and Zambetti PF (1996). Vitamin E cuts plastic taste in flexible food packaging. Converting Magazine, February, Hoffmann-La Roche Inc.

14.

Lopez-de-Dicastillo C, Pezo D, Nerín C, López G, Catalá R, Gavara R, et al. (2012) Reducing oxidation of foods through antioxidant active packaging based on ethyl vinyl alcohol and natural flavonoids. Packaging Technology and Science 25: 457–466.

15.

Nerín

(2010) Antioxidant active packaging and antioxidant edible films. In: Decker

Elias

McClements

(eds) Oxidation in foods and beverages and antioxidant applications. Cambridge: Woodhead Publishing.

16.

Nerín

Tovar

Salafranca

(2008) Behaviour of a new antioxidant active packaging versus several oxidizable model compounds. Journal of Food Engineering 84: 313–320.

17.

Palamand

Dieckmann

(1974) Autoxidation of n-hexanal. Identification and flavor properties of some products of oxidation. Journal of Agricultural and Food Chemistry 22: 503–506.

18.

Paradiso

Summo

Trani

Caponio

(2008) An effort to improve the shelf life of breakfast cereals using natural mixed tocopherols. Journal of Cereal Science 47: 322–330.

19.

Paradiso

Summo

Pasqualone

Caponio

(2009) Evaluation of different natural antioxidants as affecting volatile lipid oxidation products related to off-flavours in corn flakes. Food Chemistry 113: 543–549.

20.

Sajilata

Savitha

Singhal

Kanetkar

(2007) Scalping of flavors in packaged foods. Comprehensive Reviews in Food Science and Food Safety 6: 17–35.

21.

Sjövall

Lapveteläinen

Johansson

Kallio

(1997) Analysis of volatiles formed during oxidation of extruded oats. Journal of Agricultural and Food Chemistry 45: 4452–4455.

22.

Sumithra

Bhattacharya

(2008) Toasting of corn flakes: Product characteristics as a function of processing conditions. Journal of Food Engineering 88: 419–428.

23.

Viscidi

Dougherty

Briggs

Camire

(2004) Complex phenolic compounds reduce lipid oxidation in extruded oat cereals. Lebensmittel-Wissenschaft und-Technologie 37: 789–796.

24.

Wessling

Nielsen

Leufvén

Jägerstad

(1998) Mobility of α-tocopherol and BHT in LDPE in contact with fatty food simulants. Food Additives and Contaminants 15: 709–715.

25.

Wessling

Nielsen

Leufvén

Jägerstad

(1999) Retention of α-tocopherol in low-density polyethylene (LDPE) and polypropylene (PP) in contact with foodstuffs and food-simulating liquids. Journal of the Science of Food and Agriculture 79: 1635–1641.

26.

White

Howard

Prior

(2010) Polyphenolic composition and antioxidant capacity of extruded cranberry pomace. Journal of Agricultural and Food Chemistry 58: 4037–4042.