Determination of Critical Oxygen Level in Packages for Cooked Sliced Ham to Prevent Color Fading During Illuminated Retail Display H. LARSEN, F. WESTAD, O. SøRHEIM, AND L.H. NILSEN
T
Introduction
he color of meat products such as cooked ham is of major importance for the consumer acceptance. Hence, color stability during storage and retail display is important to the meat processing industries and the retailers (Møller and others 2003; Solomon 2004). The increased use of modified atmosphere (MA) packaging for sliced cooked products has, in general, generated problems with discoloration when the products are exposed to light in retail display cabinets. Changing from vacuum to modified atmosphere packaging has introduced headspace volume and residual oxygen in the package as new factors influencing the color of cooked meat products. Discoloration is, in particular, a problem in the Scandinavian countries, where the meat processors, due to marketing reasons, employ transparent packages for sliced meat products. Denatured nitrosylhemochrome (MbFe(II)NO), which is the pigment giving cooked meat products their pink color, is very sensitive to light when exposed to oxygen (Andersen and others 1988; Andersen and Skibsted 1992; Cornforth 1994). Light fading of the pigment is a 2-step process that includes dissociation of nitric oxide (NO) from heme groups catalyzed by light, followed by oxidation of NO and heme groups by oxygen (Cornforth 1994). What type of reaction that is degrading MbFe(II)NO upon light exposure is at present not entirely clear, but free radical processes are suggested to be involved in the degradation process (Møller and others 2002, 2005). The fading process requires both light and oxygen, and if no oxygen is present, the NO split from the heme groups by light will not be oxidized, and it can then recombine with the heme (Judge and others MS 20060036 Submitted 2/10/06, Accepted 4/09/06. Authors are with MATFORSK, Norwegian Food Research Institute, Osloveien 1, N-1430 ˚As, Norway. Author Nilsen is with Gilde Norge BA, Postbox 397 Økern, 0513 Oslo, Norway. Direct inquiries to author Larsen (E-mail: hanne.larsen@ matforsk.no)
1989; Andersen and Skibsted 1992). Even small amounts of oxygen in the headspace of the package will cause a color change from fresh pink to gray within 6 to 8 h when exposed to typical lighting conditions in the refrigerated display cabinets. Other parameters that influence the extent of photooxidation of the meat pigment is the gas-to-product-volume ratio (GP ratio), intensity and type of light exposure, type of meat product (Møller and others 2000, 2003; Jakobsen 2003), quality of the raw meat, and nitrite content (Froehlich and others 1983; Dineen and others 2000; Møller and others 2003). Thus, the amount of oxygen available in the package for pigment oxidation is crucial to the color stability of cooked meats. The amount of oxygen is influenced by the oxygen transmission rate (OTR) of the package, residual oxygen level after packaging, GP ratio, and oxygen-consuming microbial activities in the meat products during storage. In order to prevent discoloration, the OTR of the packaging material (measured on flat film) should be below 30 and 40 mL O 2 /m2 × 24 h for sliced salami and sliced bologna, respectively (Yen and others 1988; Grini and others 1992). Møller and others (2003) found that the OTR of the packaging material should be even lower for sliced cooked ham packaged in modified atmosphere, as packages with an OTR value of 10 mL O 2 /m2 × 24 h (flat material) also clearly showed discoloration in their study. The OTR of the package should preferably be measured for the finished package at the storage temperature of intended use. By measuring the OTR for the finished thermoformed and sealed package using the Ambient Oxygen Ingress Rate (AOIR) method, changes due to stretching of the material, heat seals, and possible defects created under the converting process are taken into account (Larsen and others 2000; Larsen 2004). The OTR of the packages can, by this method, also be measured at the realistic storage temperature. Even if several studies are conducted on how the above mentioned factors, separately and in combination, are affecting the color stability of cooked meats, there is still a need for more precise
S: Sensory & Nutritive Qualities of Food
ABSTRACT: The effect of packages with different oxygen transmission rates (OTR), different gas-to-product-volume (GP) ratios, and various levels of residual oxygen after packaging on the color stability of cooked ham exposed to commercial retail light conditions was studied. Sliced cooked ham was packaged in thermoformed packages with OTR of 0.04 and 0.06 mL O 2 /pkg × 24 h and GP ratios of 2.6 and 4.1. After packaging, the packages were additionally divided into groups with 4 levels of residual oxygen ranging from 0.09% to 0.46%. The packaged ham was stored in darkness at 4 ◦ C up to 33 d, and during the storage period samples were withdrawn and exposed to light for 2 d before instrumental and visual color evaluation. In order to maintain an acceptable color of this particular ham product when exposed to typical retail light conditions, the highest acceptable level of oxygen in the headspace of the packages was 0.15% oxygen at the time of illumination. This threshold level was independent of the storage time before light exposure. A residual oxygen level of below 0.15% just after packaging combined with the package with the lowest OTR (0.04 mL O 2 /pkg × 24 h) and the lowest GP ratio (2.6) was the optimal condition for maintaining the color of the tested ham product throughout the entire storage period. Keywords: color stability, cooked ham, photooxidation, residual oxygen concentration
Color fading of cooked ham . . . knowledge of how these factors interact under commercial light and temperature retail conditions during the entire shelf-life period. Under commercial conditions, the light exposure period will depend on several factors, such as the product turnover and the type of chill cabinet used in the retail stores. However, the product will very rarely be exposed to light all through the shelf-life period, which is a commonly used experimental setup seen in the literature. The objective of the present work was to study the effect of different levels of residual oxygen after packaging, 2 GP ratios, and 2 types of packaging films with different OTR on the color stability of cooked sliced ham exposed to commercial retail light conditions. Another aim in this work was to determine the maximum oxygen content in the package headspace for this particular product to avoid color fading during retail display.
Materials and Methods Product The cured, cooked meat product used for the packaging experiment was a commercial ham product produced by Gilde Fellesslakteriet (Sarpsborg, Norway). Carcasses of 6-mo-old pigs were used for the experiment. The raw materials for the ham were pork shoulder and leg muscles, with the final product containing 4% fat. The recipe consisted of meat, water, sodium chloride, glucose, sodium diphosphate, carrageenan, sodium ascorbate, sodium nitrite, and spices. After grinding of the meat through a plate with 25-mm openings, spices and curing brine were added and the mixture was tumbled for approximately 6 h before stuffing into polyamide casings with a diameter of 92 to 94 mm (Naturin GmbH & Co. KG, Germany). The hams were heated to a core temperature of 74 ◦ C, with consecutive cooling in brine to a core temperature of approximately −1 ◦ C. After cooling the hams were stored at 2 to 4 ◦ C for 6 to 24 h before slicing and packaging. The residual nitrite content in the cooked ham was 12.80 mg/kg as analyzed by Norsk Matanalyse (Oslo, Norway) according to European prestandard ENV120143:1998E (spectrophotometrically measured at 540 nm using the diazotization reaction of sulphanilamide and subsequent coupling with N-(1-naphthyl)ethylenediamine).
Packaging of product The ham was cut in slices with a thickness of approximately 0.7 mm and packaged in semirigid trays on a commercial thermoforming machine. The packages were evacuated and subsequently flushed with 100% N 2 before sealing. The size of the packages was 105 mm (width) × 160 mm (length) × 27 mm (depth). The ham was packaged in 2 different quantities, 120 or 80 g in each package, giving a GP ratio of 2.6 and 4.1, respectively. Both quantities were packaged in 2 different packages with slightly different OTR. Although both the packages had relatively low OTR values, they are denoted high- and low-OTR packages in the following. Combining the same flexible, transparent top web with 2 different semirigid, transparent bottom webs made the different packages. The main distinction of the bottom webs was different thickness of the ethylene vinyl alcohol (EVOH) barrier layer, 3 μm in the high-OTR pack-
age and 8 μm in the low-OTR package. The OTR for the flat webs, according to information from the film suppliers, were 2 mL O 2 /m2 × 24 h (23 ◦ C, 0% RH) for the top web, <3 mL O 2 /m2 × 24 h (23 ◦ C, 50% RH) for the high-OTR bottom web (high-OTR package) and 1 mL O 2 /m2 × 24 h (23 ◦ C, 50% RH) for the low-OTR bottom web (low-OTR package). After packaging the ham was transported to Matforsk for storage and analyses.
Selection and grouping of packages with different residual oxygen levels The O 2 concentration in the packages was measured at Matforsk at the day of packaging. The O 2 concentration was measured by removing 10-mL gas samples with a 10-mL Plastipak disposable syringe (Becton Dickinson SA, Madrid, Spain) with a TO-120 needle with side-hole (G W¨ohl Consulting AB, Vintrosa, Sweden). The samples were withdrawn through self-sealing Toray rubber seal patches (Part. nr. T0125; Toray Engineering Co. Ltd., Tokyo, Japan) attached to the packages. The gas samples were injected into a MOCON/Toray oxygen analyzer LC-700F (Modern Controls Inc., Minn., U.S.A.) with a zirconium oxide cell. The system accuracy of the oxygen analyzer was ±2% in the interval 0.00 to 50.00% O 2 and ±3% in the interval 0.000 to 0.500% O 2 . The oxygen analyzer was calibrated before starting the measurement of the O 2 concentration, toward air (21 mol%) in the upper part of the scale and a calibration gas containing 0.21 mol% in nitrogen (Yara A/S, Norway) in the lower part of the scale. In order to obtain the targeted (highest) levels of residual oxygen, a few mL of air was injected into some of the packages by using the disposable syringe. After measurement and adjustment of the residual oxygen in all the packages, the packages with 120 and 80 g product were divided into 4 groups with different levels of residual oxygen; see Table 1.
Storage, light exposure, and experimental setup After dividing the packages in groups with different oxygen levels, the packages were stored in darkness at 4 ◦ C for up to 33 d. Samples for oxygen analysis (O 2 concentration before light exposure, 3 replicates of each) were removed after 3, 5, 10, 15, 20, and 33 d of storage, with subsequent light exposure for 48 h before visual color evaluation. The light exposure period was selected on the basis of data supplied from another subproject in the study carried out by Østfold Research Foundation (Fredrikstad, Norway). Data from 10 retail stores showed that the ham product used in the experiment on an average was exposed to light for approximately 24 h before sale. To simulate a worst case condition and for practical reasons, an exposure time of 48 h was chosen in the experimental setup. The light exposure of the packages with ham was performed at 4 ◦ C with fluorescent light tubes placed at a distance of approximately 20 cm over the shelves. The light tubes were Osram L36W/76 Natura de luxe (Osram AS, Oslo, Norway), which are similar to the light tubes commonly used in supermarkets and recommended for lighting of cooked meat products. The light intensity was in the range 5 to 7 watt/m2 (corresponding to 1000 to 1800 lx for these light tubes). Light intensity was measured by a calibrated spectroradiometer
S: Sensory & Nutritive Qualities of Food
Table 1 --- Target and actual oxygen level in the high and low OTR packages with 120 and 80g ham Actual range of measured O 2 levels Target level % oxygen 0.10 0.20 0.30 0.40
High OTR --- 120 g
Low OTR --- 120 g
High OTR --- 80 g
Low OTR --- 80 g
0.12–0.17 0.20–0.22 0.26–0.33 0.38–0.46
0.13–0.17 0.19–0.22 0.28–0.32 0.34–0.40
0.09–0.13 0.18–0.23 0.26–0.32 0.36–0.43
0.09–0.13 0.18–0.23 0.26–0.34 0.35–0.39
Color fading of cooked ham . . . (Apogee/StellarNet Spectroradiometer with spectraWiz software, mod. SPEC-UV/PAR, Apogee Instruments Inc., Logan, Utah, U.S.A.). After 48 h of light exposure, a visual color evaluation was performed by a laboratory panel on the intact packages with ham under the fluorescent tubes in the chilling chamber. After the visual color evaluation, the O 2 concentration after light exposure was measured with subsequent opening of the package and instrumental color measurements. Microbial analysis was performed at day 0, 12, 22, and 35. The experimental setup is shown in Table 2.
Analyses of the packages and the product Measurement of oxygen concentration in the headspace of the packages. The O 2 concentration in the packages was measured after various days of storage according to Table 2. Oxygen analyses were performed before and after illumination by removing 10-mL gas samples with a syringe through self-sealing patches attached to the packages as described earlier. New packages were applied for each measurement during the storage experiment.
Visual color evaluation
Oslo, Norway). Serial 10-fold dilutions were made and 100 μL of the appropriate dilution were spread on duplicate plates on PCA (Plate Count Agar; Oxoid AS, Oslo, Norway) for enumerating total counts of bacteria and MRS Broth (Oxoid AS, Oslo, Norway) with pH adjusted to 5.7 for lactic acid bacteria. The PCA plates were incubated aerobically at 20 ◦ C for 3 d and the MRS plates aerobically at 30 ◦ C for 3 d.
Measurement of OTR of the packages OTR values of 3 replicates of the high- and low-OTR packages were measured by the AOIR method (Larsen and others 2000). The packages were flushed with nitrogen, and the O 2 concentration in the packages was measured after 1 and 5 d of storage at 4 ◦ C and approximately 60% RH outside and 100% RH (5 mL water) inside. The O 2 concentration was measured by the use of a specially designed syringe for gas sampling, and the gas sample was injected into the MOCON/Toray oxygen analyzer LC-700F as described earlier. The OTR of the packages was calculated according to the equations given in the paper by Larsen and others (2000).
A 5-member experienced laboratory panel performed the visual color evaluation. The panelists were habitual with these types of meat color measurements, and prior to the experiment they were trained on samples with different discoloration and the application of the color scale. The color of the ham in the different packages was evaluated through the transparent top web. A 5-point color scale was used including: 1 = bright pink, 2 = pink, 3 = slightly pink (still marketable), 4 = slightly gray (not marketable), and 5 = extremely gray (not marketable). The area of discoloration was assessed using a scale of 1 = none, 2 = 0 to 10%, 3 = 11% to 20%, 4 = 21% to 60%, and 5 = 61% to 100% of the light exposed area (adapted from AMSA [1991]).
On the basis of two O 2 concentration measurements in the nitrogen-flushed packages (after 1 and 5 d), the OTR of the packages was determined according to the AOIR method as described above. Additionally, by using the mathematical framework developed for the AOIR method, the O 2 concentration can be calculated (predicted) at any time in the nitrogen-flushed packages, as verified in the paper by Larsen and others (2002). The change in headspace oxygen concentration was calculated over a time period of 33 d for empty packages flushed with nitrogen and stored at 4 ◦ C.
Instrumental color measurements
Statistical analysis
∗ ∗ ∗
CIE L a b values (lightness, redness, yellowness) of the ham were measured with a Minolta Chroma Meter CR-300 (Minolta Camera Co., Osaka, Japan) with 8-mm viewing port, 20◦ viewer angle, and illuminant D 65 . Before each measurement, the instrument was calibrated against a white tile (L = 97.16, a = 0.25, and b = 2.09). The measurements were performed through a thin, transparent polyethylene film, and for each package of ham, measurements on 3 different locations were averaged. The CIE a∗ value (redness) was previously found to give the best correlation with the visual evaluation of the color of ham (Andersen and others 1988).Thus, the a∗ value was used for describing the color in this experiment.
Microbial analysis Total and lactic acid bacteria counts were performed at day 0, 12, 22, and 35 with 3 replicates of each variant at 2 oxygen levels (lowest and highest, giving a total of 6 packages from all variants each time of sampling). Ten grams of ham were mixed with 90 mL of peptone water solution in a stomacher (Colworth 400, VWR International,
Calculation of oxygen concentration in nitrogen flushed packages
Statistical analyses were performed with Minitab Release 14 (Minitab Inc., Pa, U.S.A.) and The Unscrambler ver. 9.2 (Camo Process, Oslo, Norway).
Results and Discussion Changes in headspace oxygen concentration in the packages during storage The headspace oxygen concentration in the packages was influenced by the OTR of the package, residual oxygen after packaging, and microbial activities consuming oxygen during the storage period. The interactions of these 3 phenomena in the ham packages are summarized in Figure 1. The OTR of packages without product was measured by the AOIR method at the test conditions 4 ◦ C with 5 mL water inside. The OTR values were 0.06 ± 0.00 mL O 2 /pkg × 24 h for the high-OTR package and 0.04 ± 0.00 mL O 2 /pkg × 24 h for the low-OTR package, showing that the thicker EVOH layer in the base web of the low-OTR package
Day of measurement Treatmenta High OTR --- 120 g Low OTR --- 120 g High OTR --- 80 g Low OTR --- 80 g
O 2 -levels 4 4 4 4
Analyses O 2 before light exposure O 2 after 48 h in light Color measurements Visual color evaluation Microbial analysis
0
x x
3
5
7
x
x x x
x x
10
12
x
15
17
x x x x x
20
22
x x x
33
35
x x x x x
x x x x
a High OTR --- 120 g: high OTR package with 120 g ham; Low OTR --- 120 g: low OTR package with 120 g ham; High OTR --- 80 g: high OTR package with 80 g ham; and Low OTR --- 80 g: low OTR package with 80 g ham. Three replicates were analyzed from each variant.
S: Sensory & Nutritive Qualities of Food
Table 2 --- Experimental setup
Color fading of cooked ham . . . lowered the OTR value for this package. According to the method described by Larsen and others (2002), the changes in headspace oxygen concentration were further calculated over a time period of 33 d for packages flushed with nitrogen and stored at 4 ◦ C. The calculated oxygen ingress curves are shown in Figure 1 for the highand low-OTR packages. The oxygen concentration increases more rapidly in the high-OTR package, reaching approximately 0.60% and 0.75% oxygen in the package, given an initial oxygen concentration of 0.15% (Figure 1a) and 0.3% (Figure 1b), respectively. The measured headspace oxygen concentration in the packages with ham is also shown in Figure 1. The curves are mean values averaged over all the 4 initial oxygen concentrations, showing the average changes in headspace oxygen concentration during the storage time. Figure 1a shows the average changes with 0.15% oxygen as initial oxygen concentration and Figure 1b with 0.3% as initial oxygen concentration. Figure 1a and b shows that the oxygen concentration in all the packages with ham is lower than in the empty packages and decreases with time due to oxygen consumption by microbial activity. Figure 2 shows that the total bacteria increased during storage, while headspace oxygen decreased (Figure 1a), indicating that with bacterial growth, there was a reduction in headspace oxygen levels. The oxygen concentration decreased in all the packages after approximately 15 to 20 d of storage, and the microbial load increased from approximately 104 to 107 colony forming units (CFU) per gram in the similar time interval. Similar growth curves as recorded in this work are reported by other researchers with maximum
S: Sensory & Nutritive Qualities of Food
Figure 1 --- Calculated changes in oxygen concentration in empty packages during 33 d and measured average changes in oxygen concentration in packages with 120 or 80 g (GP ratio 2.6 and 4.1, respectively) sliced ham during 33 d of dark storage. Figure 1a with 0.15% as initial oxygen concentration and Figure 1b with 0.3 % as initial oxygen concentration.
microbial levels of approximately 108 CFU per gram after 20 to 30 d of storage (Grini and others 1992; Houben and Gerris 1998; Møller and others 2003). The growth curves for lactic acid bacteria were almost equal (not shown) to the total bacteria curves, suggesting that most of the bacteria were different types of lactic acid bacteria. Lactic acid bacteria were also found to be dominating at the end of the storage period in other studies with sliced bologna and ham (Ahvenainen and others 1989; Grini and others 1992; Houben and Gerris 1998). Figure 1 also shows that the oxygen reduction tends to be highest in the packages with 120-g product compared to packages with 80 g, especially for the high-OTR package, probably due to a higher total microbial activity consuming oxygen in the package. The oxygen concentration in the high-OTR package was significantly higher (P < 0.005) than in the low-OTR package after 10, 15, and 20 d of storage, as also observed by the statistical analysis in Table 3. This finding was in agreement with the higher OTR value of the high-OTR package compared to the low-OTR package.
Correlation between oxygen concentration in the packages and color A visually acceptable product (score 3 = slightly pink) corresponded to a CIE a∗ -value (redness) of 8 (correlation coefficient = −0.91, P < 0.005) for this particular ham product. Samples with a∗ -values lower than 8 had various degrees of gray color and were considered as unfit for sale. Figure 3 shows the correlation between the oxygen concentration in the packages before light exposure and the a∗ -value measured after 48 h of light exposure. Figure 3 shows a high negative correlation (linear correlation coefficient = −0.87, P < 0.005) between the oxygen concentration and the a∗ -value. Both a linear and a nonlinear correlation line are drawn in Figure 3, and the nonlinear curve seems to give the best fit. Figure 3 shows that as the oxygen concentration increased, the a∗ -value decreased. At oxygen concentrations below approximately 0.14%, most of the samples had higher a∗ -values than 8 and were considered as acceptable (samples within the marked area in Figure 3). At oxygen concentrations higher than 0.14%, the samples had different degrees of gray color. Samples with a∗ -values below approximately 4 were extremely gray. Figure 4 shows the correlation between the oxygen concentration in the packages before light exposure and the visual evaluation of color after 48 h of light exposure. Figure 4 also shows a high correlation (linear correlation coefficient = 0.89, P < 0.005) between the oxygen concentration and the visual evaluation. Both a linear and a nonlinear correlation line are drawn for visual color in Figure 4, and
Figure 2 --- Total bacteria counts on the ham in the different packages during 35 d of storage
Color fading of cooked ham . . . Table 3 --- Effect of film permeability and GP ratio on oxygen concentration and color after various days of storage Days of dark storage Factor
3
5
Effect of film and GP ratio on oxygen concentration before light exposure Package permeability (high or low) NS NS GP ratio 2.6 and 4.1 NS NS
10
15
20
33
+++ NS
+++ NS
+++ NS
NS NS
Days of dark storage prior to illumination Factor
3
5
10
Effect of film and GP ratio on ham color after dark storage and 48 h of light exposure Package permeability (high or low) NS NS +++ GP ratio 2.6 and 4.1 NS NS +
15
20
33
+++ +
+ +
NS NS
NS: not significant. +++: P < 0.005. ++: P < 0.01. +: P < 0.05.
as for the a∗ -value, the nonlinear curve also seemed to give the best fit for the visual color. At oxygen concentrations below approximately 0.17%, the majority of the samples had acceptable color with visual color values from 1 (=bright pink) to 3 points (=slightly pink, see samples within the marked area in Figure 4). At oxygen levels higher than 0.17%, most of the samples had different degrees of gray color with visual color values from 3.5 to 5 (=extremely gray). The instrumental color analysis and the visual color analysis indicated slightly different threshold oxygen levels in the packages before one could record a shift from acceptable to unacceptable color. The difference was very small, and was considered as insignificant from a practical point of view. By summing up the results from both the instrumental analysis and the visual evaluation, a highest acceptable level of 0.15% oxygen in the headspace of the packages could be set for this particular ham product and lighting conditions to prevent color fading when exposed to light. This threshold level was independent of the storage time before light exposure. This finding is in accordance with results obtained by Møller and others (2000), who found that the limit lay between 0.1% and 0.5% oxygen in the headspace of the packages for the sliced cooked ham used in their experiment.
Figure 3 --- Correlation between oxygen concentration in the packages before light exposure and a∗ -value. The solid-drawn line shows a linear relationship (r = −0.87), and the dotted line a nonlinear relationship.
Figure 4 --- Correlation between oxygen concentration in the packages before light exposure and visual color evaluation. The solid-drawn line shows a linear relationship (r = 0.89), and the dotted line a nonlinear relationship.
S: Sensory & Nutritive Qualities of Food
quantities) on the oxygen level before light exposure and the ham color after 48 h of light exposure. Table 3 shows that the oxygen concentration in the packages was significantly different at d 10, 15, and 20, where the high-OTR package had the highest O 2 levels (see Figure 1 as well). The different GP ratios had no significant effect on the oxygen level. The lower oxygen concentration in the low-OTR package also affected the color. The ham in the low-OTR package had higher a∗ -values than the high-OTR package after 10, 15, and 20 d of dark storage prior to illumination. Even if the different GP ratios did not show any significant effect on the oxygen concentration, the GP ratio was found to significantly influence the color of the ham after 10, 15, and 20 d. At these points of time, the average a∗ -values for the packages with GP ratios of 4.1 and 2.6 were approximately 4.75 and 6.50, respectively. The average a∗ -values were accordingly below the acceptance limit of 8 for both GP ratios, but the ham in the packages with GP ratios of 4.1 had an overall more gray color due to a more severe oxidation. Our findings are in accordance with the results obtained by Møller and others (2003). They stated that a low GP ratio should be used to maintain a high a∗ -value, but the GP ratio became less important at increased oxygen levels. Above 0.85% oxygen there was sufficient oxygen available for pigment degradation regardless of the GP Statistical analysis of the effect of films and GP ratio ratio. Analysis of variance (ANOVA) was performed analyzing the effect A higher GP ratio could on the other hand be partially compenof different films and GP ratios (2.6 and 4.1 due to different product sated for by a lower oxygen concentration in the packages. The
Color fading of cooked ham . . . oxygen analyses performed in this experiment, given in Table 1, showed that the residual oxygen level in general was slightly lower in the packages with the highest GP ratio, probably due to better evacuation of these packages. Both the GP ratio and the oxygen concentration account for the total amount of oxygen available for photooxidation (Møller and others 2003).
Oxygen level in the packages before and after light exposure Andersen and Skibsted (1992) and Møller and others (2002, 2005) have shown that there is a linear relationship between oxygen pressure and light-induced degradation expressed as quantum yields. In aqueous solutions the stoichiometry for photooxidation was even found to be higher than 1:1, indicating that a different reaction mechanism is operating for the photooxidation than for the thermal oxidation. Andersen and Skibsted (1992) describe the mechanism of photooxidation in 2 steps: MbNO + hv → MbNO∗
(1)
where an electronically excited state of MbNO is produced by absorption of a photon, which reacts in a bimolecular process with a ground-state oxygen (triplet state: 3 O 2 ): MbNO∗ + 3 O2 → MMb + other products
(2)
giving metmyoglobin (MMb) and other products. At present it is not entirely clear what type of reaction is degrading MbFe(II)NO upon light exposure, but free radical processes are suggested to be involved in the degradation process (Møller and others 2002, 2005). The equations above show that oxygen is consumed in the photooxidative reaction. Figure 5 shows how the oxygen concentration rapidly decreased in 1 set of packages with 4 different initial oxygen levels during 48 h of light exposure.
S: Sensory & Nutritive Qualities of Food
Figure 5 --- Headspace oxygen levels (mL) in packages before and after 48 h of light exposure (high-OTR package, 120 g, 10 d of dark storage before oxygen measurement and light exposure). The oxygen concentration (%) in the packages before light exposure and the a∗ -values are also shown for each level and the dark reference sample.
The highest oxygen consumption was observed in the packages with the highest oxygen level before light exposure, and the color of the ham was most severely degraded in these packages as well as shown in Figure 5. The ham in the packages stored in darkness did not consume oxygen. Rather, the oxygen concentration slightly increased due to oxygen ingress through the package. With oxygen levels lower than approximately 0.15% in the packages (corresponding to 44 and 46 mL oxygen in the packages with GP ratio of 2.6 and 4.1, respectively), the total amount of oxygen molecules available for photooxidation was not sufficient to give a significant visible color degradation for the ham product and package used in this experiment. In order to avoid color fading of cooked meat products, the meat processors can choose 2 main strategies; eliminating oxygen or light. This work has demonstrated that in order to eliminate oxygen to the essential low level, the packaging material must have a very high oxygen barrier and the running of the packaging machine must be performed by well-trained operators to maintain a low residual oxygen level after packaging. The oxygen measurements on the day of packaging showed a relatively large variation in the oxygen level in the packages produced under equal conditions. The variation in residual oxygen level was most probably due to different effectiveness on the 4 parallel lines of the packaging machine in vacuuming and flushing the packages. Good maintenance routines for the packaging machines and training of the operators are therefore important to secure optimal running of the machines. Andersen and others (1988) found for vacuum-packed sliced cooked ham that an initial dark storage period of 4 d prior to light exposure improved the color stability due to efficient depletion of oxygen in the packages during the 4 d of storage. Oxygen depletion did not occur until after 15 to 20 d in the packages in this experiment. The product is usually exposed to light in the retail store before 15 d, indicating that an oxygen level above the critical limit would cause discoloration in many of the packages. A more effective strategy would be to lower the residual oxygen level in the packages at the time of packaging. Oxygen can also be eliminated by the use of oxygen absorbers, as demonstrated by, for example, Grini and others (1992) in packages for sliced bologna. The largest single category of products currently employing oxygen absorber technology is sliced meat products such as smoked and cured meats (Solomon 2004). Different absorber solutions are available, depending on the consumer acceptance in different countries. The use of oxygen absorbers is not yet common in Northern Europe due to legal limitations, negative consumer attitudes, and the cost of the absorber. Eliminating light exposure is another strategy to reduce the problem with discoloration of sliced meat products. The most commonly used packaging solution for sliced meat products in Southern Europe is a printed top web with light barrier combined with a transparent thermoformed bottom web. The printed top web is turned toward the fluorescent light tubes and the consumer, but the consumer can evaluate the product appearance just by turning the package upside down. A higher headspace oxygen concentration could probably be accepted by using this solution, and an optimal running of the packaging machine would be less critical.
I
Conclusion
n order to maintain an acceptable color of this particular ham product when exposed to commercial retail light conditions, the highest acceptable level of oxygen in the headspace of the packages was 0.15% oxygen. This threshold level was independent of the storage time before light exposure. A residual oxygen level of below 0.15% just after packaging combined with the package with the
Color fading of cooked ham . . .
Acknowledgments The authors thank Gilde Fellesslakteriet and Gilde Vest for the support of product and use of packaging machinery, and Wihuri OY Wipak (Nastola, Finland) and Bemis Valkeakoski Oy (Valkeakoski, Finland) for the supply of film materials used in the project. We would also like to thank Aud Espedal, Karin Solgaard, Anne-Kari Arnesen, and Janina Berg for skilful assistance performing the oxygen analyses, visual evaluation, and microbial analyses. The work was performed within a collaboration project between Matforsk, Østfold Research Foundation, Gilde Norge BA, Prior Hærland AS, and Fjordland AS. The Research Council of Norway provided financial support.
References
Ahvenainen R, Skytt¨a E, Kivikataja RL. 1989. Factors affecting the shelf-life of gas- and vacuum packed cooked meat products. Part I: Sliced ham. Lebensm Wiss Technol 22:391–8. [AMSA] American Meat Science Assn. 1991. Guidelines for meat color evaluation. In: 44th Annual Reciprocal Meat Conference; 1991 June 9-12; Manhattan, Kans. Chicago, Ill: National Livestock and Meat Board. p 3–17. Andersen HJ, Skibsted LH. 1992. Kinetics and mechanism of thermal oxidation and
photooxidation of nitrosylmyoglobin in aqueous solution. J Agric Food Chem 40:1741–50. Andersen HJ, Bertelsen G, Boegh-Soerensen L, Shek CK, Skibsted LH. 1988. Effect of light and packaging conditions on the color stability of sliced ham. Meat Sci 22:283– 92. Cornforth D. 1994. Color—its basis and importance. In: Pearson AM, Dutson TR, editors. Quality attributes and their measurement in meat, poultry and fish products – Vol. 9. London: Blackie Academic & Professional. p 34–78. Dineen NM, Kerry JP, Lynch PB, Buckley DJ, Morrissey PA, Arendt EK. 2000. Reduced nitrite levels and dietary α-tocopherol acetate supplementation: effects on the colour and oxidative stability of cooked hams. Meat Sci 55:475–82. Froehlich DA, Gullett EA, Usborne WR. 1983. Effect of nitrite and salt on the color, flavor and overall acceptability of ham. J Food Sci 48:152–4. Grini JA, Sorheim O, Nissen H. 1992. The effect of packaging materials and oxygen on the colour stability of sliced bologna. Packag Technol Sci 16:249–57. Houben JH, Gerris CVM. 1998. Effect of the dietary supplementation with vitamin E on colour stability of packaged, sliced pasteurized ham. Meat Sci 50:421–8. Jakobsen M. 2003. Optimizing MAP of meat through modelling; modelling colour stability and changes in headspace gas composition during storage. [DPhil dissertation]. Copenhagen: Food Chemistry, Dept. of Dairy and Food Science, The Royal Veterinary and Agricultural Univ. p 57. ISBN 87-7611-033-8. Judge MD, Aberle ED, Forrest JD, Hedrick HB, Merkel RA. 1989. Principles of meat processing. In: Judge MD, Aberle ED, Forrest JD, Hedrick HB, Merkel RA, editors. Principles of Meat Science. 2nd ed. Dubuque, Iowa: Kendall/Hunt Publishing Co. p 135–174. Larsen H. 2004. Oxygen transmission rates of packages at ambient, chill and freezing temperatures measured by the AOIR method. Packag Technol Sci 17:187–92. Larsen H, Kohler A, Magnus EM. 2000. Ambient oxygen ingress rate method—an alternative method to Ox-Tran for measuring oxygen transmission rate of whole packages. Packag Technol Sci 13:233–41. Møller JKS, Jensen JS, Olsen MB, Skibsted LH, Bertelsen G. 2000. Effect of residual oxygen on colour stability during chill storage of sliced, pasteurised ham packaged in modified atmosphere. Meat Sci 54:399–405. Møller JKS, Bertelsen G, Skibsted LH. 2002. Photooxidation of nitrosylmyoglobin at low oxygen pressure. Quantum yields and reaction stoechiometries. Meat Sci 60:421–5. Møller JKS, Jakobsen M, Weber CJ, Martinussen T, Skibsted LH, Bertelsen G. 2003. Optimisation of colour stability of cured ham during packaging and retail display by a multifactorial design. Meat Sci 63:169–75. Møller JKS, Nannerup L, Skibsted LH. 2005. Effect of carbon dioxide on autoxidation and photooxidation of nitrosylmyoglobin. Meat Sci 69:71–8. Solomon J. 2004. Eliminating oxygen. Meat Poultry 9:38–41. Yen JR, Brown RB, Dick RL, Acton JC. 1988. Oxygen transmission rate of packaging films and light exposure effects on the color stability of vacuum packaged dry salami. J Food Sci 53:1043–6.
S: Sensory & Nutritive Qualities of Food
lowest OTR (0.04 compared with 0.06 mL O 2 /pkg × 24 h) maintained the color of the tested ham product throughout the entire storage period. However, if the residual oxygen level just after packaging was above approximately 0.15%, applying the package with the lowest OTR did not prevent color degradation when the ham was exposed to light during the first 10 d of storage. The oxygen level in most of the packages decreased at the end of the storage time due to microbial activity, and the color fading of the ham when exposed to light was hence less pronounced at the end of the storage time. The color of the ham in packages with the GP ratio of 2.6 was significantly better than in packages with a ratio of 4.1 after 10, 15, and 20 d of storage.
