Highlights
- •
-
In-package cold plasma allows to effectively control spoilage organisms in chicken breast.
- •
-
Cold plasma has minimal effect on colour, pH and water holding capacity of poultry meat.
- •
-
Plasma treatment allowed to extend the microbial shelf-life by approximately 6 days under refrigerated storage.
- •
-
In-package cold plasma is a viable alternative to chemical disinfection of poultry surfaces.
Abstract
Atmospheric cold plasma (ACP) is a promising non-thermal technology for controlling food spoilage. In this study, ACP treatment at 100 kV for 1, 3 and 5 min was applied to chicken breast samples. Approximately 2 log CFU/g reduction in natural microflora of chicken was achieved within 5 min of treatment and 24 h of storage. The observed reduction was attributed to the reactive oxygen and nitrogen species in cold plasma. For shelf-life study, control and ACP treated samples (100 kV for 5 min) were analysed for the population of mesophiles, psychrotrophs and Enterobacteriaceae as well as sample colour and pH over a storage period of 24 days. On day 24, the population of mesophiles, psychrotrophs and Enterobacteriaceae in treated chicken was respectively 1.5, 1.4 and 0.5 log lower than the control. These results suggest that in-package ACP is an effective technology to extend the shelf-life of poultry products.
Keywords
Industrial Relevance
Fresh poultry products including chicken breast meat are highly perishable. Atmospheric cold plasma is a promising non-thermal technology for controlling food spoilage and extending shelf-life. This work demonstrates the efficacy of in-package cold plasma treatment in decontaminating chicken breast surface and extending the shelf-life. Being an energy efficient and chemical free decontamination intervention, in-package cold plasma is a promising alternative traditional chemical disinfection of poultry surfaces.
1. Introduction
The popularity of poultry meat has been steadily increasing around the world over the last few years. Of the various poultry cuts, the consumption of chicken breast fillets is becoming quite popular due to the relatively low cost of production of a high protein food with low fat content and the convenience of a ‘boneless’ meat portion. Due to the perishable nature of chicken meat, the poultry industry remains concerned about shelf-life extension methods for poultry products. The microflora in poultry meat starts off on its surface and their population at any time-point depends on the initial counts and storage conditions. It is well established that this bacterial growth and its activity on the surface of the product are the main causes of the spoilage and unacceptability of chicken products among consumers, resulting in industrial economic losses (Fernández-Pan, Carrión-Granda, & Maté, 2014). In general, whole poultry carcasses are known to carry lower microbial populations than cut-up poultry. An increase in the mean surface counts of microflora from 3.30 log CFU/cm2 to 3.81 log CFU/cm2 after cutting-up, and to 4.08 log CFU/cm2 after packaging was reported long back (May, 1962). To decrease the microbial populations, poultry processors employ a variety of sanitizing treatments, including the use of organic acids (e.g. acetic, lactic, citric, and succinic), chlorine, chlorine dioxide, trisodium phosphate, and acidified sodium chlorite (Keener, Bashor, Curtis, Sheldon, & Kathariou, 2004). However, the increasing consumer concern about the use of chemical based (or chemical ‘sounding’) sanitizers is forcing poultry processors to look for alternative decontamination methods.
Among the emerging innovative antimicrobial technologies, cold plasma is being extensively explored for its surface decontamination capabilities. Cold plasma is essentially an ionised state of a gas achieved by exposing the gas (air or any gas mixture) to very high electric field strengths. The ionised gaseous chemical species include the positive and negative ions, radicals, electrons, excited and neutral molecules, and quanta of electromagnetic radiation (i.e. ultraviolet and visible light) (Fukuda, Kawasaki, & Izawa, 2019). As an example, humid air plasma would contain reactive oxygen and nitrogen species (RONS) such as ozone, singlet oxygen, peroxide, and several types of nitrogen oxides (NxOy). Most of the RONS possess strong antimicrobial activity. These RONS species render cold plasma into an excellent surface decontamination technology.
In an earlier study, Dirks et al. (2012) had shown that plasma application to the surface of raw chicken breast allowed to decrease the natural microflora by 0.85 log units. Similarly, Lee et al. (2016) reported about 1.2 log CFU units reduction in the population of total aerobic bacteria on chicken breast after 5 min of plasma treatment using a flexible thin layer DBD plasma set-up. Recently, Misra and Jo (2017) had reviewed the application potential of cold plasma technology in the meat industry with great attention to the details for effective decontamination. Cold plasma has several advantages with respect to poultry processing, including the ability to combine with other hurdle approaches such as controlled release packaging (Pankaj et al., 2017), essential oil treatments (Chouliara, Karatapanis, Savvaidis, & Kontominas, 2007), refrigeration, and modified atmospheric packaging (Misra et al., 2014). Moreover, cold plasma could overcome the limitations of some approaches such as the requirement for use of large quantities of essential oils to observe reasonable control of the microflora population.
Among the various cold plasma configurations, in-package cold plasma technology is a distinct approach involving the ionization of gas contained in a sealed package in presence of the food product. For in-package plasma treatment, the product is first sealed inside a plastic (or occasionally glass) packaging in air or a modified gas, and the package is exposed to a strong electric field. The ionised gaseous species have a high diffusivity and long lifetime, thus ensuring a uniform treatment of the product. However, being quasi-stable, the reactive species eventually recombine to form the original gas. Thus, the process begins with a known gas mixture (or air) and ends with the same gas mixture after few hours, while effectively eliminating a significant population of the spoilage microorganisms (Misra, Yepez, Xu, & Keener, 2019).
In this work, we have studied the in-package cold treatment of chicken breast in air at a high voltage of 100 kV. The objectives of this study were to (1) evaluate the efficacy of high voltage in-package cold plasma treatment at 100 kV for various treatment times in decreasing the natural microflora in chicken breasts; (2) measure the changes in quality parameters, viz. colour, pH, lipid oxidation, and water holding capacity; (3) study the long-term effects of in-package cold plasma treatment on the native microflora population and quality of chicken breasts stored under refrigerated conditions.
2. Materials and methods
2.1. Experimental set-up
Cylindrical chicken breast samples were obtained via coring-out sections from dorsal side of chicken breasts using a sterile hollow stainless-steel cylinder (height ~ 28 mm, diameter ~ 23 mm) with sharp edges. For all experiments, fresh chicken breast sample (10 g) with no specificity to dorsal or ventral side facing the top was placed at the centre of a polypropylene rigid package with dimensions 168 mm × 120 mm × 28 mm (ArtBin® box). The rigid package was further placed inside a secondary package comprising of a high barrier polypropylene film (Cryovac®). Prior to heat sealing, the package was flushed with dry air (<5% relative humidity) for 2 min. The samples intended for use as control, as well as those for plasma treatment were packaged under the same condition. Samples meant for microbiological analysis were plasma treated separately from those for quality. This method prevented any chances of contamination of samples, that could possibly result from handling during quality testing, that in turn could have affected results of microbiological analysis.
A detailed, labelled schematic of the experimental set-up is provided in Fig. 1. The plasma source employed in for the experiments comprised of a dielectric barrier discharge (DBD) with a step-up transformer providing the power. The step-up transformer operated with an input voltage of 120 V at 60 Hz frequency that could be varied using a voltage regulator. The output from the secondary winding of the step-up transformer was connected to two circular aluminium disc electrodes (outer diameter = 152 mm) separated by 42 mm. The dielectric barriers comprised of 10 mm plexiglass under the powered (high voltage) electrode and a 4 mm polypropylene sheet above the ground electrode. It may be noted that the polypropylene package itself acted as a dielectric. The dielectric layers prevent a transition of the filamentary discharge into an arc, thereby ensuring homogeneity of the plasma treatment. The electrodes were pre-cooled to 5 °C prior to sample treatment using two ice packs. The electrode temperatures were maintained close to 5 °C throughout the plasma treatment process, as measured using an infrared thermometer with an accuracy of ±1 °C (Fisherbrand™ Traceable™, Fisher Scientific, USA), by leaving the icepacks in contact with the electrodes. For the experiments looking at effect of time, the samples were treated at 100 kV for 0, 1, 3, and 5 min durations. The system consumed an average power of 233 ± 5 W for operating at 100 kV, as measured using a power meter (P4460 Kill A Watt, P3 International; procured from Grainger, USA). After plasma treatment, all samples were stored in sealed condition at 4 °C for 24 h prior to microbial recovery and quality measurements.
Fig. 1. Experimental set-up used for plasma treatment of the chicken samples. (colour online).
2.2. Optical emission spectroscopy (OES)
The arrangement for OES is shown in Fig. 1. The light emitted from the electrical discharge in the package was captured using a 5 mm diameter collimating lens directed towards a solarized optical fiber with a core diameter of 1000 μm. The distance from the collimating lenses to the edge of the box containing sample was set at 7 cm. The optical fiber directed the light to a computer controlled, custom-built Ocean Optics spectrometer (Ocean Optics, Inc., Florida, USA) with 0.2 nm/pixel resolution. The grating spectrometer had a 10 μm slit width with an optical resolution of 0.88 nm. The spectrometer was pre-calibrated by the manufacturer using a mercury‑argon atomic line source. Measurements were carried out over the wavelength window of 200–800 nm of the electromagnetic spectrum. The integration time for spectral recording was set to 2 s and an average of 10 spectrum was recorded, thereby maximizing the signal to noise ratio. Each spectrum was corrected by the software (Ocean View, Ocean Optics, FL, USA) for dark current and the background noise via subtraction, and the averaged spectra were reported. The characteristic molecular and atomic transitions associated with the spectral bands and lines were identified using published reports and National Institute of Standards and Technology (NIST) database (Misra et al., 2014; Misra, Keener, Bourke, Mosnier, & Cullen, 2014).
2.3. Microbiological analysis
The objective of the microbiological analysis was to quantify the effect of cold plasma treatment on the background bacterial microflora population in the chicken samples. For bacterial enumeration, the chicken sample was aseptically removed from each package and placed into a sterile stomacher bag. Ninety millilitres of sterile 0.1% peptone solution was added to each bag and the sample was pummelled for 3 min using a Stomacher 400 Laboratory Blender (Seward, Worthington, UK) operating at 230 rpm. The resulting suspension was serially diluted (10-fold) in 0.1% peptone and 0.1-ml aliquots were surface plated on appropriate agar media for each microbial group. For the aerobic plate count samples were surface plated on plate count agar (PCA) and incubated at 35 °C for 48 h for mesophiles and at 7 °C for 7 days for psychrotrophs. The Enterobacteriaceae were enumerated by first pour plating samples in tryptic soy agar supplemented with 0.6% yeast extract (TSAYE), followed by holding the TSAYE plates for 1 h at ambient temperature (22 ± 1 °C). Each solidified TSAYE plate was overlaid with violet red bile (VRB) agar and incubated at 35 °C for 24 h. After the incubation periods, the colonies were manually counted, and the bacterial populations were computed using the respective dilution factors. The bacterial populations were reported as log colony forming units (CFU) per gram.
2.4. Analysis of quality
2.4.1. Colour measurement
The colour of the control and plasma treated chicken samples was quantified using a Hunter L⁎-a⁎-b⁎ colorimeter (Colour Quest XE Hunter Lab, Northants, U.K.). The observer angle was set at 10° and the light source was a standard D65 halogen lamp. Prior to evaluation of the samples, the instrument was calibrated using standard white (L⁎ = 93.97, a⁎ = 0.88 and b⁎ = 1.21) and green (L⁎ = 56.23, a⁎ = 21.85, b⁎ = 8.31) tiles. Colour measurement for each sample was done in triplicate (dorsal, ventral, and lateral sides each) and average values were taken for reporting purpose.
2.4.2. Analysis of pH
A 1.0 g of cold plasma treated, or untreated sample was homogenised at 16,000 rpm with 9.0 ml spectroscopy grade water (Sigma Aldrich, Product No. 270733-4L) for 30 s. The pH of the sample was measured using an Orion Dual Star pH/SE meter. The pH meter was calibrated with the standard buffer solutions (pH 4.00, 7.00, and 10.00) at room temperature prior to measurements.
2.4.3. Water holding capacity (WHC)
The WHC of meats is measured via the amount of free water released by the meat after physical pressure or force is exerted upon it. The WHC was determined using a centrifugation method adopted from Zhang and Barbut (2005) and Wardlaw, McCaskill, and Acton (1973). About 5 g of finely chopped chicken breast sample were mixed with 8 ml of cold 0.6 M sodium chloride solution and homogenised for 1 min at 8000 rpm (UltraTurrax T25 homogenizer) followed by incubation at 4 °C for 30 min. After incubation, the meat slurry was again stirred for 1 min followed by centrifugation at 16,000 g for 20 min. The supernatant layer was collected and measured by volume. The amount of added solution retained by the meat is reported as the water holding capacity in ml per 100 g meat.
2.4.4. Lipid oxidation
The method for lipid oxidation measurement by 2-Thiobarbituric Acid Reactive Substances (TBARS) assay was adapted from Botsoglou et al. (1994) and Beltran, Pla, Yuste, and Mor-Mur (2003), with slight modifications. A 2-g chicken breast meat sample was transferred into a 50 ml centrifuge tube with 8 ml 5% aqueous trichloroacetic acid (TCA) and 5 ml 0.8% butylated hydroxytoluene (BHT) in hexane. The mixture was then homogenised at 16,000 rpm for 30 s. The homogenate was centrifuged for 7 min at 15,000 g, and the upper hexane layer was discarded. The 2.5 ml aliquot from the lower aqueous layer was mixed with 1.5 ml of 0.8% aqueous thiobarbituric acid (TBA) followed by 25 min incubation at 80 °C. The tubes were then cooled in running tap water for 10 min and then stabilised at room temperature for 30 min. Subsequently, the absorbance was read at 521 nm using a spectrophotometer (Synergy HT, BioTek Instruments, Winooski, VT). The TBA content was expressed as mg of MDA per kg of chicken using a calibration curve prepared for MDA concentration versus absorbance (see Supplementary information, SI. 1), which was interpolated via linear regression.
2.5. Storage study
Based on the results of the effects of treatment time, a separate storage study was carried out for chicken breast samples treated at 100 kV for 5 min. For the storage studies, both control and plasma treated samples were stored aerobically under refrigerated conditions (4 °C) until further analysis. The stored samples were sampled on days 0 (i.e. after 24 h of in-pack storage), 3, 6, 9, 12, 15, 18, 21 and 24 for microbial enumeration (mesophiles, psychrotrophs, and Enterobacteriaceae), and quality studies. Due to time constraints and the labour involved, only colour and pH were the quality characteristics measured for the storage study.
The evolutions of microbial populations over the storage period was fitted to an empirical logistic bacterial growth model and the growth rate parameter obtained was used for comparing the growth of microbial populations in control versus plasma treated chicken samples. The logistic growth model is described by Eq. (1) (Pankaj, Misra, & Cullen, 2013).(1)
where N(t) is the population at any time t, N0 (log CFU/g) is the initial population, Nasymp (log CFU/g) is the population number approached at the stationary phase (which is an asymptote), k (h−1) is the rate constant, and tc (h) is a marker of the inflection point. One could in practice, employ any of the several models available for microbial growth modelling (Peleg & Corradini, 2011), and not necessarily the logistic type of growth model. We chose the logistic model for the relative simplicity of this model and ease of parameter interpretation. The fitted models were used for predicting the time to reach the cut-off values for each microbial group.
2.6. Statistical analysis
The experiments evaluating the effects of treatment time were repeated thrice and each set was replicated five times, thereby resulting in 15 measurements. For the storage studies, all experiments were performed in triplicates. Statistical analysis and plotting of data were carried out through scripts written in the open-source R programming software (R foundation for statistical computing, Vienna, Austria). Analysis of variance (ANOVA) was used for mean comparison and the data was represented as mean value ± standard deviation. Statistical significance was evaluated at p ≤ 0.05. The fitting of experimental microflora evolution data during storage to the logistic model was done through nonlinear regression using the Levenberg–Marquardt algorithm for least—squares optimization, and the goodness of fit was evaluated via the coefficient of regression (Radj2) and the Root Mean Square Error (RMSE) statistics.
3. Results and discussion
3.1. Optical emission spectroscopy
The chemistry of plasma discharges in air has been widely recognised to be very complex, with the involvement of over 75 species, that interact via over 1000 reaction schemes (Misra et al., 2018). Furthermore, these reactions are multiscale in nature, occurring over a range of length and time scales. Emission spectroscopy of plasma allows study of the nature of the excited species in plasma. The emission spectra for the air plasma in presence of the chicken samples is shown in Fig. 2. The spectrum reveals the presence of strong emissions in the wavelength range of 315–405 nm corresponding to electronic transitions from nitrogen second positive system, N2(C-B) and first negative system, N2+ (B-X). The band heads of the N2(C3Πu → B3Πg) second positive system were recorded around 336.9 nm, 357.3 nm, 380.0 nm and 405.4 nm, while the spectral emission of the nitrogen mono-positive ion N2(B2∑u+ → X2∑g+) was recorded at 390.6 nm and 427 nm with relatively low intensities. The intense spectral signatures indicated the occurrence of energetic collisions of electrons with molecular nitrogen in air. During the electrical discharge process, electrons acquire enough energy (temperature) to ionize dominant air molecules (N2 and O2) and form the excited N2* molecules:
Fig. 2. Optical emission spectrum of the plasma discharge inside the sealed package in presence of chicken breast. (colour online).
Fast quenching of the excited N2* molecules with molecular oxygen is one of the sources of atomic oxygen:
However, the relatively lower intensities of OES peaks associated with oxygen is due to self-quenching of O(3P) in the air plasma (Walsh, Liu, Iza, Rong, & Kong, 2010):
The singlet oxygen formed in plasma also undergoes self-recombination or react with ozone (O3) to produce molecular oxygen:
Ozone is a well-known antibacterial species that forms in the plasma discharges, and has also been measured earlier using chemical analysis and optical absorption spectroscopy at significant concentrations with the DBD system employed in our experiments (Mahnot, Mahanta, Keener, & Misra, 2019; Moiseev et al., 2014).
3.2. Microbial inactivation versus treatment time
Microorganisms are often classified according to their growth temperature as either thermophiles (growth temperature: >50 °C), mesophiles (growth temperature: from 20 °C to 50 °C), or psychrophiles (growth temperature: <20 °C). With respect to the current study, evaluations of the numbers of mesophiles and psychrotrophs are important with regard to the spoilage of chicken breast meat intended for human consumption. The Enterobacteriaceae are a group of Gram-negative bacteria whose population is often considered a hygiene indicator during poultry processing (Chouliara et al., 2007). With respect to mesophiles, psychrotrophs, and Enterobacteriaceae, the mean counts were 4.5 log cfu/g, 4.6 log CFU/g, and 3 log CFU/g, respectively, for untreated samples (0 min and 24 h storage), indicative of good quality chicken meat (see Fig. 3). After 5 min of treatment at 100 kV and 24 h in-package storage, the mean populations of mesophiles, psychrotrophs, and Enterobacteriaceae were found to have decreased to 2.6 log cfu/g, 3 log CFU/g, and 1.3 log CFU/g, respectively. Thus, 5 min of plasma treatment allowed to decrease the native microflora by approximately 1.5 log CFU/g. This decrease is higher than the 0.85 log units reported by Dirks et al. (2012), possibly due to rapid production of large concentrations of RONS. However, it is comparable to the 1.5 log units decrease reported by Kronn et al. (2015), who employed a modified atmosphere gas blend (65% O2, 30% CO2, 5% N2) for in-package treatment with a similar DBD plasma system. The treatment time was found to have a significant effect (p < 0.05) on the decrease in population for all three microbial classes. This rapid decrease in population was due to the generation of large concentrations of RONS species inside the package with longer treatment times, which act on the microbial cells resulting in their oxidation and leakage (Misra & Jo, 2017). The increased levels of microbial inactivation with increasing treatment time could also be in-part due to increased susceptibility to inactivation by plasma RONS as a result of the drop in pH (discussed later in section 3.4). It is worthwhile mentioning that our results contradict the reports of Zhuang et al. (2019), who observed that the antimicrobial effectiveness against food-borne pathogens is not influenced by cold plasma treatment time (from 60 s to 300 s) at 70 kV. This is most likely due to the drastically different plasma chemical dynamics at 70 kV versus 100 kV.
Fig. 3. Evolution of the population of (A) aerobic mesophiles, (B) psychrotrophs, and (C) Enterobacteriaceae in chicken breast samples (n = 15) subjected to in-package cold plasma treatments at 100 kV, 60 Hz frequency in air. Populations are presented as log transformations to base 10. The numeric values above each ‘jittered’ time group of data represent the p-value for pairwise mean comparison. The p-value for overall significance of the differences obtained via analysis of variance (ANOVA) is also included. The relative population values are colour coded, and lines of best fit obtained from linear regression are included for ease of visualization only. (colour online).
3.3. Changes in colour parameters
The change in colour parameters of the chicken breast samples as a function of treatment is summarised in Fig. 4. The L* (Lightness), a* (redness), b* (yellowness) values of plasma-treated chicken breasts were found to be closely related to the untreated control samples, with no statistically significant difference between them (p > 0.05). This insignificant difference persisted treatments up to 5 min followed by in-pack storage for 24 h. The results were consistent for all the three colour parameters (L*, a*, and b*). Our results are in agreement with those reported by Wang, Zhuang, Lawrence, and Zhang (2018), who reported insignificant changes in the all the colour parameters of chicken fillets subjected to cold plasma treatments using a dielectric barrier discharge source operating at 80 kV. However, in an earlier study, plasma treatment using a surface barrier discharge was found to result in greenness and enhanced lightness in chicken breast samples (Lee et al., 2016). A change in colour of other types of meat samples exposed to plasma discharges from different plasma sources (microwave and surface barrier discharge) has been reported in some studies (Fröhling et al., 2012; Jayasena et al., 2015). Such differences are likely related to the plasma chemistry associated with the type of discharge systems used and their operating parameters. For example, Jayasena et al. (2015) employed a surface barrier discharge that is known to have a very different plasma chemistry as compared to a volumetric discharge used in our experiments (Moiseev et al., 2014; Park, Choe, & Jo, 2018). Overall, it becomes clear that the plasma source employed in our study did not lead to any significant change in the colour of the chicken breast samples after treatment.
Fig. 4. Evolution of the colour parameters, viz. (A) lightness, L*, (B) redness-greenness, a*, and (C) blueness-yellowness, b* of chicken breast samples (n = 15) subjected to in-package cold plasma treatments at 100 kV, 60 Hz frequency in air. The numeric values above each ‘jittered’ time group of data represent the p-value for pairwise mean comparison. The p-value for overall significance of the differences obtained via analysis of variance (ANOVA) is also included. The relative values are colour coded, and lines of best fit obtained from linear regression are included for ease of visualization only. (colour online).
We further carried out treatments for extended times of 15 min (while the experiments were only done for 5 min), to record the maximum increase in temperature. For 15 min of operation, the plasma source employed in our study does not lead to an increase in temperature by >7 °C – 8 °C. That said, even a 5 °C to 6 °C rise in temperature of the samples could cause some discoloration of the chicken, especially an increase in the lightness. This was confirmed during our preliminary trials prior to design of the final experimental plan for this work. Recently, Zhuang et al. (2019) have also observed this effect when using a similar plasma source for treatment of chicken breast samples, with no reported cooling of the electrodes. Therefore, preventing a rise in sample temperature throughout the plasma process, via cooling of the electrodes is important for maintaining the quality of chicken breasts. It is worthwhile mentioning that such cooling of electrodes may not very critical when treating other kind of products such as fruits or vegetables (Misra et al., 2014).
3.4. Changes in chemical quality
3.4.1. Change in pH
The pH of the control and plasma treated samples are shown in Fig. 5a. The pH of control chicken breast samples was found to vary between 5.6 and 6.1 units. An overall decrease in pH of the chicken samples was observed for all treatment times. However, the pH drop was statistically significant (p < 0.05) between 1 min and 3 min of treatment, but not between 3 min and 5 min of treatment. This is likely linked to the plasma chemistry, where much of the NxOy form during the initial 1 min to 3 min of treatment, following which they are dominated by production of or conversion to other reactive species. A decrease in pH of chicken and other meats following exposure to plasma discharge has been reported in several other studies (Kim, Yong, Park, Choe, & Jo, 2013; Rothrock Jr. et al., 2017). Additionally, a significant drop in the pH of aqueous samples exposed to similar type of DBD plasma had revealed a considerable drop in pH (Misra, Keener, Bourke, & Cullen, 2015). In our earlier work, post-discharge, we had confirmed the presence of nitrous species (nitrate and nitrite) in the gas phase (Mahnot, Mahanta, Keener, & Misra, 2019). The decrease in pH is commonly attributed to the formation of very low concentration of nitric and nitrous acid from the dissolution of NxOy species formed in the gaseous phase into the water covering the muscles-.
Fig. 5. Evolution of the chemical quality parameters, viz. (A) pH, (B) Water Holding Capacity, WHC (mL/100 g), and (C) malonaldehyde content (mg/kg), of chicken breast samples (n = 15) subjected to in-package cold plasma treatments at 100 kV, 60 Hz frequency in air. The numeric values above each ‘jittered’ group of data represent the p-value for pairwise mean comparison. The p-value for overall significance of the differences obtained via analysis of variance (ANOVA) is also included. The relative values are colour coded, and lines of best fit obtained from linear regression are included for ease of visualization only. (colour online).
In some studies, however, it has been reported that no change in the pH was recorded in meat samples even after plasma treatment; e.g. for beef loin (Bauer et al., 2017) and bacon (Kim et al., 2011). Such differences are likely linked to the plasma chemistry, the surface moisture content, the humidity in gas, and the buffering ability of the muscle type.
3.4.2. Change in WHC
The 70–75% of water in the muscles of live poultry is bound primarily to the muscle proteins. This ability of muscle proteins to intracellularly bind about 90% of the water is referred to as WHC (Honikel, 1987). The WHC of the control chicken breast was found to be close to 24 ml/100 g (see Fig. 5b). This value is in agreement with that reported by Zhuang and Savage (2012). Plasma treatments up to 3 min were found to result in insignificant changes (p > 0.05) in the WHC of the muscle cells. However, the treatment for 5 min was found to result in a significant (p < 0.05) loss of WHC reaching about 17 ml/100 g. It is well-known that a accelerated drop in pH of lean muscles results in a decrease in their WHC, due to a decrease in the space in the myofibril compartment (Huff-Lonergan & Lonergan, 2005). Further, a denaturation of proteins on surface of the chicken samples cannot be overruled. Such denaturation will also result in a decrease in the WHC of the muscles.
3.4.3. Lipid oxidation
Upon oxidation, lipids result in the formation of peroxides, which further decompose to secondary oxidation products, including malonaldehyde (MDA). Therefore, MDA is used as an indicator of lipid oxidation and deterioration in meat and meat products. The TBARS assay involves the condensation of one molecule of malonaldehyde with two molecules of 2-thiobarbituric acid under heated acidic conditions to form a pink chromogen, which was spectrophotometrically measured in the assay. The results of TBARS assay indicated that the difference in mean MDA content was insignificant (p > 0.05) between control and plasma treated samples at all treatment times (Fig. 5c). Our results are in agreement with the findings of Lee et al. (2016) for surface barrier discharge plasma treated chicken breast. In a recent review of the effects of cold plasma on lipid oxidation in meat and meat products, it was reported that chicken breast was more stable to plasma-led oxidation than red meats (Gavahian, Chu, Mousavi Khaneghah, Barba, & Misra, 2018). Such differences can be attributed to the very low-fat content of lean chicken breast as compared to red meat (pork or beef).
3.5. Microbial population during storage
The evolution of microbial populations during the storage period are summarised in Fig. 7, and the summary of model fitting is provided in Table 1. The model fitting was found to be adequate with high R2 (adj) values and low RMSE values. Following plasma treatment, the population of all the three microbial groups were significantly decreased by 1 to 1.5 log CFU/g. This was also evident from the modelling parameter N0 (Table 1). All the three groups of micro-organisms, viz. mesophiles, Enterobacteriaceae, and psychrotrophs grew at a slower rate in plasma treated chicken than control, as can be observed from Fig. 6a, b and c), respectively. The model fitting also confirmed that the growth rate parameter k (h−1) was lower for plasma treated microbial group as compared to control for all three groups (see Table 1). Towards the end of storage study, the microbial populations for control and plasma treated samples converged towards nearly the same population levels of ~10 log CFU/g, with the plasma treated samples still exhibiting lower values (see Nasymp parameter in Table 1). The lag phase was found to be significantly extended for mesophiles, and psychrophiles after plasma treatment by >100 h (tc parameter). However, the Enterobacteriaceae were found to have similar lag phase values for control as well as plasma treated samples, thus indicating a quicker recovery from plasma-led injury. However, despite the similar lag phase periods, the population of Enterobacteriaceae remained lower in plasma treated samples vis-à-vis control, throughout the duration of storage study (Fig. 6b).
Table 1. Summary of model parameters for different microbial groups and the statistical parameters of non-linear regression.
| Sample | N0 (log CFU/g) | Nasymp (log CFU/g) | tc (h) | k (h−1) | R2 (adj) | RMSE | |
|---|---|---|---|---|---|---|---|
| Mesophiles | Control | 3.5 ± 0.4 | 10.5 ± 0.1 | 137.6 ± 11.3 | 0.016 ± 0.002 | 0.98 | 0.32 |
| Plasma | 2.0 ± 0.3 | 9.7 ± 0.2 | 243.7 ± 10.5 | 0.010 ± 0.001 | 0.98 | 0.31 | |
| Enterobacteriaceae | Control | 2.8 ± 0.1 | 10.6 ± 0.1 | 246.7 ± 3.2 | 0.040 ± 0.003 | 0.99 | 0.32 |
| Plasma | 1.3 ± 0.2 | 8.9 ± 0.1 | 254.0 ± 5.9 | 0.017 ± 0.001 | 0.98 | 0.34 | |
| Psychrotrophs | Control | 4.1 ± 0.2 | 10.0 ± 0.1 | 137.2 ± 5.3 | 0.030 ± 0.004 | 0.98 | 0.31 |
| Plasma | 2.6 ± 0.2 | 9.6 ± 0.1 | 249.5 ± 7.4 | 0.014 ± 0.001 | 0.98 | 0.32 |
Fig. 6. Evolution of the microbial populations in control (■) and plasma treated (●) chicken stored at 4 °C: (A) Mesophiles, (B) Enterobacteriaceae, (C) Psychrotrophs. The symbols correspond to experimental data, whereas solid lines are curves obtained from fitting of the logistic model via non-linear regression. Populations are log transformations to base 10. (colour online).
When used together, refrigeration and modified atmosphere packaging has been shown to ensure a minimum shelf-life period of 8 days (Chouliara et al., 2007). The recommended cut-off population limits of Enterobacteriaceae and aerobic mesophiles for acceptability of chicken breasts is 7 log CFU/g and 6 log CFU/g, respectively (Chouliara et al., 2007; Fernández-Pan et al., 2014). While no cut-off limits are reported in literature for psychrotrophs, we chose a value of 7 log on a conservative side. In this study, the Enterobacteriaceae population was found to reach 7 log CFU/g on day 10, the psychrotroph population on day 5 and the mesophiles population was found to reach 6 log CFU/g on day 4 for the control samples. For the plasma treated samples, the Enterobacteriaceae and psychrotroph population were found to reach 7 log CFU/g on day 13 and day 12, respectively. The mesophiles population was found to reach 6 log CFU/g on day 10 for plasma treated samples. Thus, by using a simple principle of minimum extension time, cold plasma treatments in air effectively enabled extending the microbial shelf-life by approximately 6 additional days at 4 °C storage temperature. That said, a complete shelf-life evaluation of plasma treated chicken for dynamic conditions should also be confirmed through evaluation of bacterial growth at different temperatures (Mahnot, Mahanta, Farkas, Keener, & Misra, 2019)).
3.6. Evolution of colour and pH during storage
The variations in colour parameters of the chicken breast samples over the storage period are summarised in supplementary file (supplementary information SI 2). The difference in the mean values of green-red (a*) and blue-yellow (b*) parameters were found to be non-significant (p > 0.05) between the control and plasma treated group over the 24 days storage period. However, the b* value was observed to be significantly different (p < 0.05) only on day 15. This is likely due to a natural variability in the colour of those specific samples, as the b* values were found to be insignificant for the subsequent days. Unlike the a* and b* values, the lightness parameter was found to be significantly (p < 0.05) different for control and plasma treated samples after the 9th day of storage. On the 24th day, the difference was again found to be insignificant (p > 0.05). The decrease in lightness (L*) of the control could be attributed to the formation of slime on the surface and spoilage caused by the activity of higher microbial population as compared to plasma treated samples.
The changes in pH of the control and plasma treated samples during storage are summarised in Fig. 7. The difference in pH of the control and plasma treated chicken samples was found to be statistically insignificant until 9th day (p > 0.05). From the 12th day onwards the pH of control was found to increase, which was significant (p < 0.05) as compared to plasma treated samples. The increase in pH of control samples can be correlated with a rapid increase in the population of micro-organisms, and their proteolytic activity results in the formation of basic compounds (Vinci & Antonelli, 2002).
Fig. 7. Change in pH of control and plasma treated chicken breast samples stored at 4 °C. The significance of the difference between mean values is indicated above each day (*’s indicate that the difference was significant, while ns refers to non-significant difference between the mean values). (colour online).
4. Conclusions
Our work reveals that in-package cold plasma treatment at 100 kV can considerably inhibit growth of spoilage micro-organisms on lean poultry meat surfaces without drastically altering the quality parameters. At least 1.5 log CFU/g reduction in the population of mesophiles, Enterobacteriaceae, and psychrotrophs was confirmed after 5 min of plasma treatment at 100 kV and in-pack storage for 24 h. The variations in the quality parameters between control and plasma treated chicken breast was practically negligible. A storage study under refrigeration revealed that plasma treatment allows to extend the shelf-life of the chicken breasts by 6 days. These effects render in-package cold plasma treatment as a promising technology for extending the shelf-life of poultry meat.
As a next step, exploration of the use of modified gases is encouraged to further decrease the microflora population and increase the shelf- life of poultry meat. In addition, a detailed study of the volatile profile of chicken meat subjected to cold plasma treatment should also be carried out to assess the effects of reactive species in cold plasma.
Declaration of Competing Interest
None.
Acknowledgements
Authors acknowledge Dr. Buddhi Lamsal, Iowa State University for providing access to the temperature-controlled centrifuge.
Appendix A. Supplementary data
Supplementary material.
References
- Bauer et al., 2017
-
A. Bauer, Y. Ni, S. Bauer, P. Paulsen, M. Modic, J.L. Walsh, F.J. SmuldersThe effects of atmospheric pressure cold plasma treatment on microbiological, physical-chemical and sensory characteristics of vacuum packaged beef loinMeat Science, 128 (2017), pp. 77-87, 10.1016/j.meatsci.2017.02.003
Huff-Lonergan and Lonergan, 2005
