Efficacy of cold plasma in producing Salmonella-free duck eggs: effects on physical characteristics, lipid oxidation, and fatty acid profile

Abstract

Concerns related to foodborne pathogens necessitate the decontamination of avian eggs. Conventional decontamination methods, such as egg washing, usually use health-threatening chemicals (e.g. chlorine). Hence, innovative chemical-free decontamination approaches are interesting for the food industry, especially to decontaminate commonly Salmonella infected products such as duck eggs. The present study is the first attempt to evaluate the effectiveness of cold plasma against Salmonella enterica inoculated on the surface of duck eggshell. In this regard, Salmonella-contaminated duck eggs were treated by arc plasma for 10, 20, 30, and 40 s. The bacteria count, eggshell strength, color, pH, Haugh unit, acid value (AV), thiobarbituric acid reactive substances (TBARS), and fatty acid profile of the plasma-treated samples were then compared with those of untreated sample. According to the results, all the plasma treatments significantly decreased the Salmonella population and longer treatment times enhanced the bactericidal effects of plasma. A maximum bacterial reduction of 4.1 log cycle was observed when plasma was applied for 40 s. Furthermore, plasma treatments did not deteriorate the quality parameters of eggs such as eggshell strength, eggshell color, yolk color, Haugh unit, AV, and TBARS. These findings introduced arc plasma as an emerging tool for improving the safety of duck eggs with good potential for industrial application.

Electronic supplementary material

The online version of this article (10.1007/s13197-019-03996-z) contains supplementary material, which is available to authorized users.

Introduction

The eggs of various birds, such as hen and duck, are good sources of macro- and micro-nutrients and play a major role in the human diet. However, the consumption of avian eggs can be concerning as they are prone to be polluted with infectious Salmonella species (Wan et al. 2017). Salmonella enterica is a predominant Salmonella serotype and is one of the main safety concerns associated with the consumption of eggs and egg-containing foods (Wan et al. 2017). It was reported that about three-fourths of the Salmonella outbreaks were caused by consumption of eggs and egg-based products (CDC 2014). Also, Salmonella was responsible for about one-third of the 790 outbreaks with a laboratory-confirmed illness (CDC 2014). Washing eggs with chemicals, such as chlorine (sodium hypochlorite) solutions, is the most common decontamination practice of egg (Al-Ajeeli et al. 2016). However, recent studies discouraged this decontamination method due to its potential negative health consequences (Ölmez and Kretzschmar 2009; Shenga et al. 2010). For example, it was revealed that chlorine reacts with organic compounds of the egg and generates carcinogenic by-products of halogenated disinfection (Ölmez and Kretzschmar 2009). Besides, washing eggs with chemicals produces a huge amount of chemically contaminated waste which can be an environmental concern. Therefore, emerging chemical-free approaches, such as cold plasma, have been recently proposed for egg decontamination (Georgescu et al. 2017; Wan et al. 2017; Dasan et al. 2018).

The term “plasma” alludes to a quasi-neutral ionized gas, consists of reactive species, superoxide, peroxide, hydroxyl radicals, photons, free electrons, ions, and atoms with bactericidal (Thirumdas et al. 2016; Gavahian and Khaneghah 2019) and oxidative effects (Gavahian et al. 2018a). Researchers highlighted the potential applications of cold plasma in the food (Gavahian and Cullen 2019) and poultry industry (e.g. hen egg decontamination) (Gavahian et al. 2019). However, the relatively high capital cost of plasma equipment makes it more suitable for processing pricier commodities, such as duck eggs, with regards to commercial production. It was reported that the possibility of Salmonella contamination in duck eggs can be higher than that of hen eggs (Advisory Committee on the Microbiological Safety of Food 2016) and the consumption of duck eggs was the reason for several cases of Salmonella-caused infections and outbreaks around the world (Owen et al. 2016). This could be related to the physical structure of duck eggshell and duck farming system. For example, previous studies confirmed that the eggshell of the duck egg is more porous than that of the hen egg which provides a shelter-like structure and increases the possibility of Salmonella survival during egg decontamination processes (Kaewmanee et al. 2013; Quan and Benjakul 2019). Besides, the production of duck eggs is often associated with backyard breeders, multiple age production, manual collection of eggs, insufficient biosecurity measures, and the lack of all-in/all-out production system and, therefore, is less structured than those of industrial chicken farms. Moreover, the absence of safety regulations and Salmonella control program for duck farms is a common issue in many regions of the world (Advisory Committee on the Microbiological Safety of Food 2016).

Therefore, the feasibility of duck egg decontamination through emerging decontamination techniques, such as cold plasma, needs to be explored to provide new information for farmers, food processors, and researchers. This study, for the first time, has been conducted to explore the effects of 10, 20, 30, and 40 s of arc plasma treatment on microbial safety of duck eggs and some quality parameters of the product including acid value (AV), thiobarbituric acid reactive substances (TBARS), fatty acid profile, and eggshell color.

Materials and methods

Bacterial strain and preparation of inoculum

Salmonella enterica Subsp. enterica (BCRC 10747) was supported by Bioresource Collection and Research Center of Food Industry Research and Development Institute (BCRC of FIRDI, Hsinchu, Taiwan). The pure strain was isolated from the freeze-dried tube according to the streak-plate method using nutrient agar (DifcoTM, United States) at 37 °C for 24 h. The isolated colonies were then harvested by a sterile loop and suspended in 20 mL of nutrient broth in a 50 mL sterile plastic tube. The tube was capped and incubated for 24 h at 37 °C while shaking at the speed of 150 rpm. Afterward, the bacteria were pelleted by centrifugation (Kubota-3700, Kubota, Japan) at 3500 rpm for 10 min while the temperature was kept constant at 4 °C. Finally, the resulted pellet was suspended in 2 mL of nutrient broth to reach a final cell concentration of 10¹⁰ CFU mL⁻¹.

Duck eggs and spot inoculation

Duck eggs were obtained from Wang Mingyuan duck farm (Hsinchu, Taiwan). The weight of eggs (before and after plasma treatments) was measured in triplicates using an FX-2000i digital balance (A&D, Japan) with the precision of 0.01 g. Before inoculation, eggs were thoroughly washed by reverse osmosis (RO) water and their surface was sterilized by swabbing with 70% ethanol (Taiwan Tobacco and Liquor Corporation, Taiwan) and then complete immersion in 70% ethanol for 1 min. Decontaminated eggs were put in the horizontal direction in a biosafety cabinet for air drying. Afterward, the eggs were gently spotted (totally 5 spots for each sample) by depositing 100 μl S. enterica inoculum on the eggshell within a selected square-shaped area with the approximate dimensions of 2 × 2 cm. The eggs remained in their position for 10 min to prevent inoculum run-off. Afterward, the eggs were air-dried for 1.5 h in the biosafety cabinet at ambient temperature to ensure that the bacterial cells are attached to the eggshell (Wan et al. 2017). All the experiments were replicated four times and the inoculated eggs were immediately used for plasma treatments.

Plasma equipment and treatments

A laboratory-scale arc atmospheric pressure plasma (APPL-10k, Taiwan) was used for duck egg treatment in this study. This plasma device consists of a control panel, a voltage transformer, an air pump, and two independent plasma-generating chambers. The plasma gas was produced through one of the plasma-generating chambers of the device which consists of two electrodes. The working gas was the common atmospheric air (room air) with the relative humidity of 65% (Crecer-HD-75 Hygrometer, Japan). The technical parameters in the present study were as follow: Working distance: 65 mm; Pressure: 101 kPa; Treatment time: 0, 10, 20, 30, and 40 s; Temperature: 20–49 °C; Plasma voltage: 12,000 V; I: 1.9–2 A; Maximum plasma power: 24,000 W (all the electrical specifications were reported according to the measurement equipment supplied by APPL Company, Taiwan).

Briefly, artificially contaminated eggs (one egg in each treatment) were horizontally placed in close contact with the generated plasma gas for 0, 10, 20, 30, and 40 s (Fig. 1). 40 s was selected as the longest plasma exposure time according to the temperature–time data obtained in the preliminary study (Figure S1, supplementary materials) to make sure that the egg temperature remains below 50 °C during the plasma process. The temperature of the egg surface was recorded during plasma treatment using an infrared-based p384 thermography camera (Ching Hsing Computer-Tech Ltd, Taiwan) along with IRM-P384-19 software (Ching Hsing Computer-Tech Ltd, Taiwan). The camera was mounted on a Vel-Flo 9 PH368 tripod (Velbon, Japan) in the distance of about 0.5 m from the plasma device. In addition, a TES 1306 thermometer (TES Electric Electronic Corporation, Taiwan) was used to double-check the temperature of eggshells before and after plasma treatment in triplicate. All the plasma treatments were applied with three replications and plasma-treated eggs along with control samples were subjected to microbiological and physicochemical assays.

Fig. 1.

Schematic representation of steps involved in microbiological investigations in the present study (a), schematic representation of Salmonella inoculation and the decontamination process of the artificially decontaminated duck egg by arc plasma (the inoculation points are highlighted in this diagram with a marker for a better illustration of the process) (b), and the graphical representation of the observations in the present study (c)

Microbiological study

Negative control (disinfected eggs that were neither inoculated nor treated by plasma), positive control (inoculated and untreated eggs), and plasma-treated (the inoculated eggs that were subjected to plasma) samples were investigated for the recovery of the microbial population. The S. enterica on the inoculum area of the eggs were recovered by a BPW 3M Swab-Sampler with buffered peptone water (3M™, United States). The obtained suspensions were shaken gently for 2 min and serially diluted with sterile water. Afterward, 2 mL of the diluted solution was pipetted into a 3M Petrifilm Salmonella Express Plates (3M™, United States). After incubating at 37 °C for 24 h, these samples were added to a confirmation disk (3M™, United States) and incubated at 37 °C for 4 h. The resulted blue colonies were confirmed as Salmonella and were counted. The microbiological study of the duck egg, including spot inoculation and bacteria recovery, is presented visually in Fig. 1.

Physicochemical analysis of eggs

The physicochemical properties of plasma-treated eggs, including eggshell color, eggshell strength the pH of albumen and yolk, yolk color, Haugh unit, and the chemical composition of yolk oil, were assessed and compared to those of control (untreated) sample. All the physicochemical experiments were performed in triplicate.

Eggshells strength, thickness, and color

The strengths and thickness of eggshells were assessed using a digital egg tester (DET-6000, Nabel Co. Ltd., Japan). Besides, the visual colors of plasma-treated eggshells were compared to that of the untreated sample (Gavahian et al. 2012). Also, the eggshells of untreated and plasma-treated samples (with no physical damage) were subjected to the CM-5 laboratory colorimeter (Konica Minolta Sensing, Japan) to evaluate L^* (lightness), a^* (greenness-redness), and b^* (blueness yellowness) values. The color intensity (B value) was then calculated according to the Eq. (1) (Gavahian et al. 2018b).

$$\text{B}=\sqrt{a^{\ast 2}+b^{\ast 2}}$$ 1

where B, a*, and b* represent the color intensity, greenness-redness, and blueness-yellowness values, respectively.

Furthermore, the overall color difference (ΔE), i.e., the difference between the color of plasma-treated samples and the reference sample (untreated egg), was calculated according to the Eq. (2) (Cserhalmi et al. 2006).

$$\mathrm{\Delta }{E}^{\ast }=\sqrt{{\left(\mathrm{\Delta }{L}^{\ast }\right)}^2+{\left(\mathrm{\Delta }{a}^{\ast }\right)}^2+{\left(\mathrm{\Delta }{b}^{\ast }\right)}^2}$$ 2

where ΔL^*, Δa^*, and Δb^* are the difference between L^*, a^*, and b^* values of a plasma-treated sample and the control (untreated) sample, respectively.

The internal quality of eggs

The physical parameters (yolk color, albumen pH, yolk pH, yolk height, and Haugh unit) and chemical parameters (AV, TBARS, and fatty acid composition of yolk oil) of duck eggs were assessed and compared to those of untreated eggs.

Physical parameters

Duck egg samples were cracked and yolk color, yolk height, and Haugh unit were measured using a digital egg tester (DET-6000, Nabel Co. Ltd., Japan). The pH of the egg albumen and yolk were measured using a pH-510 laboratory pH meter (Eutech Instruments, Singapore) with a glass electrode probe (GB-900E) when the samples were stirred by an IKA C-MAG HS7 digital magnetic stirrer (Staufen, Germany). The pH meter was calibrated through a two-point calibration using 10.0 and 7.0 buffers before each experiment and all the tests were performed in triplicate.

Chemical parameters

The effects of plasma process on AV, TBARS, and fatty acid profiles of the yolks were assessed and compared with those of untreated eggs. In this regard, the yolk oil was extracted according to the method described by Paraskevopoulou and Kiosseoglou (1994). Briefly, the liquid yolks were lyophilized and mixed with hexane (1:10 w v⁻¹), grounded for 10 s in an Oster 6640 blender (Oster, United States), and extracted under continuous stirring for 4 h using an IKA C-MAG HS7 digital magnetic stirrer (Staufen, Germany). The yolk-solvent mixture was then filtered under vacuum using Whatman filter paper grade 4 (Whatman, United Kingdom). The filtrates were then rotary-evaporated (Buchi Rotavapor RII, Flawil, Switzerland) at 40 °C to volatilize the solvent and to concentrate the oil. The extracted oil was then analyzed for fatty acid composition, AV, and TBARS.

The fatty acid profile of the extracted oils from egg yolk was determined in triplicate according to the AOCS official methods (Ce-1b-89 method) (Firestone 1994). Fatty acid methyl esters were made from the oil sample by direct transesterification using 1 N NaOH in methanol. Afterward, a Gas Chromatography coupled with Flame Ionization Detector (GC-FID) (Agilent Series, 6890 Series, United States) was used to separate and quantify these fatty acid methyl esters. Briefly, 1 mL of the yolk oil was weighed in a screw-capped Pyrex culture tube and 1 mL of heptadecanoic acid (C17) (Catalog number: H3500; Purity: 98%; Concentration: 1 mg mL⁻¹) (Sigma-Aldrich, United States) was added as the internal standard. Afterward, 1 mL of 1 N sodium hydroxide in methanol was added to the tube, vortexed for 30 s, and incubated at 80 °C for 15 min. The tube was then cooled under tap water followed by the addition of 1 mL of 14% BF3 in methanol reagent (Merck, Germany), vortexed for 30 s, and incubated for 15 min at 110 °C. Afterward, 1 mL of n-Hexane (Merck, Germany) was added to the cooled tube and vortexed for 1 min. Then, 6 mL of a saturated sodium chloride solution (Merck, Germany) was added and the tube was held for separation after 5 s of slight shaking. The separated n-Hexane layer from the aqueous layer was further dehydrated under anhydrous sodium sulfate (Merck, Germany) and transferred to a clean amber glass screw cap vial. The dehydrated n-Hexane layer, which contained fatty acid methyl esters, was filtered through 0.22 µm polyvinyl difluoride (PVDF) syringe filter and injected into the GC device. The CP7489 column (100 m × 0.25 mm × 0.2 µm) was used for the GC analysis. The injector temperature was set at 250 °C, the carrier gas was pure nitrogen with the flow rate of 1 mL min⁻¹, and the split ratio was 1:40. The oven program was as follow: 80–220 °C with the rate of 4 °C min⁻¹ (hold for 5.0 min) and then increased to 240 °C (hold for 10 min). FID was used to detect the separated fatty acids and was operated at 270 °C using H₂ and zero air at the flow rates of 30 mL min⁻¹ and 400 mL min⁻¹, respectively. The injection volume was 1 µL. Supelco 37 Component Fatty Acid Methyl Esters Mixture (Sigma Cat. No. 47885-U) was used as the reference standard to quantify the fatty acids presented in the oil sample through the area percentage calculation. The results were then expressed as the percentage of fatty acids in the injected oil. Internal standard and Standard Reference Material (SRM-1544) (NIST, United States) were used for analytical quality assurance.

The AV of yolk oils was studied according to AOCS-Cd 3d-63official method (Firestone 1994) using a KOH-ethanol solution. KOH and ethanol were purchased from Merck (Darmstadt, Germany) and Taiwan Tobacco and Liquor Corporation (Taipei, Taiwan), respectively. Phenolphthalein (Merck, Germany) was used as the indicator to determine the titration endpoint and the AV was expressed as mg KOH per g of the oil sample.

Thiobarbituric acid reaction method was employed to determine the levels of lipid peroxidation (Gavahian et al. 2013). Briefly, 1.5 g of the extracted oils were mixed with 0.5 N trichloroacetic acid to reach the fixed volume of 5 mL and vortexed for 30 min (CM-1000 high-speed mixer, Eyela, Japan) to precipitate the proteins. Following centrifugation, draw the water layer and filtered through 0.22 µm PE syringe filter, mixed with an equal volume of 0.67% thiobarbituric acid in a boiling water bath for 40 min. Afterward, the sample was cooled to the room temperature and the color produced by the reaction of thiobarbituric acid and malondialdehyde was measured at 532 nm using an infinite M200 pro spectrophotometer (Tecan, Switzerland). Measurements were performed in triplicates and expressed as microgram of malonaldehyde per milliliter of yolk oil (μg MD mL⁻¹).

Statistical analysis

A total number of 444 eggs were used in this study and all the plasma experiments were performed in triplicates (148 eggs were used in each replicate). All the physicochemical and microbiological tests were repeated for three times. SPSS Statistics V. 23.0 (IBM, United States) was employed to perform the statistical analysis. The ANOVA test was carried out to identify the effects of plasma treatment on physicochemical and microbiological characteristics of duck eggs. The differences between the means of the studied parameters were compared with Post Hoc-Duncan test at a confidence interval of 0.95.

Results and discussion

The results indicated that plasma treatments slightly affected some physicochemical proprieties of the duck eggs. On the other hand, the weight of all the duck eggs (untreated and plasma-treated) was in the range of 63.55–66.21 g and was not affected by plasma treatments (Table 1).

Table 1.

The effects of plasma treatments on egg weight and some physical properties of eggshell

	Egg weight (g)	Eggshell strength (N)	Eggshell thickness (mm)	L*	a*	b*	B	ΔE*	Visual color
Control	66.21 ± 4.79a	49.28 ± 4.03a	0.45 ± 0.04a	66.13 ± 0.83a	− 4.60 ± 0.75a	4.22 ± 0.28a	6.28 ± 0.61a	–	Light green
10 s	65.12 ± 4.46a	48.25 ± 6.10a	0.46 ± 0.04a	66.20 ± 0.59a	− 4.58 ± 0.68a	4.27 ± 0.24a	6.28 ± 0.58a	0.39 ± 0.13a	Light green
20 s	64.89 ± 5.15a	48.79 ± 3.79a	0.46 ± 0.04a	66.23 ± 0.71a	− 4.60 ± 0.73a	4.29 ± 0.29a	6.31 ± 0.59a	0.32 ± 0.10a	Light green
30 s	65.18 ± 4.89a	49.43 ± 5.40a	0.45 ± 0.04a	66.08 ± 0.75a	− 4.55 ± 0.77a	4.20 ± 0.36a	6.22 ± 0.68a	0.40 ± 0.08a	Light green
40 s	63.55 ± 5.37a	48.71 ± 5.03a	0.45 ± 0.05a	66.02 ± 0.83a	− 4.54 ± 0.79a	4.30 ± 0.15a	6.28 ± 0.59a	0.39 ± 0.07a	Light green

*Data are presented as Means ± Standard deviation. Data within a column with the same lowercase letters are not significantly different at p > 0.05. Control: untreated egg; 10 s: plasma-treated eggs for 10 s; 20 s: plasma-treated eggs for 20 s; 30 s: plasma-treated eggs for 30 s; 40 s: plasma-treated eggs for 40 s; L*: lightness; a*: greenness-redness; b*: blueness yellowness; B: color intensity; ΔE: overall color difference (as compared to control sample)

Temperature variations during plasma treatments

The thermographs of duck eggs before and after 40 s of plasma treatment along with the variations in the eggshell temperature over treatment time are presented in Fig. 2. According to the results, 40 s of arc plasma treatment increased the surface temperature of eggs from 27.1 ± 1.2 °C to 48.7 ± 7.8 °C (Fig. 2c). This temperature increase is relatively high and limits the maximum plasma treatment time of the egg samples (Khani et al. 2017). Many of the commercially available plasma systems suffer from some design limitations for the treatment of food materials. For example, dielectric barrier discharge plasma, which can perform the plasma treatment at room temperature, is usually suitable for the treatment of small and thin objects as disturbances in plasma formation can be observed while treating large samples such as egg (Khani et al. 2017). In this case, expanding the electrode gap reduces the electric field, produces non-ionized gases, and decreases the efficacy of the process. Arc plasma systems that are characterized by generating high-density and high-energy electrons at relatively low temperatures (< 80 °C) have been previously suggested for the treatment of bulky objects (Khani et al. 2017). According to Khani et al. (2017), arc plasma treatment increased the temperature of tomato samples to above 40 °C which was in line with the findings of the present study. Similarly, it was reported that an arc plasma treatment increased the temperature of hem seed to about 50 °C (Sera et al. 2017). Furthermore, Baier et al. (2015) explained that elevated temperatures, i.e. above 60 °C, can be expected in the arc plasma treatment, depending on process parameters such as exposure time and the distance of the sample form plasma generation zone (Baier et al. 2015). On the other hand, the design of the arc plasma system makes it a potential candidate for industrial applications as it can process bulky food materials (e.g. duck eggs) in continuous mode (e.g. by using several plasma-generating chambers on the top of a belt conveyor which carries the food materials at a predefined speed). Further up-scaled investigations are required to assess the possibility of application of various types of plasma systems in the food industry.

Fig. 2.

Effects of forty seconds plasma treatment on the temperature of the duck egg: thermograph of the eggs before (a) and after (b) plasma treatment along with variations in temperature of selected points during plasma process (c). TP refers to the temperature of target points (TP1: the targeted point at the left side of the egg; TP2: the targeted point at the right side of the egg; TP3: the targeted point in the middle/front the egg; TP4: the targeted point on the upper side of the egg)

Decontamination effects of plasma treatment on the eggshell

Figure 3 presents the results of Salmonella inactivation during plasma treatment. The results indicate that all the plasma treatments significantly reduced the bacteria count of duck eggs. A short plasma treatment, i.e., 10 s, reduced the Salmonella population of the contaminated (untreated) eggs from 8.58 ± 0.24 to 6.29 ± 0.13 CFU cm⁻². According to the data, longer exposure times resulted in further bacteria reductions (Fig. 3). For example, 20 s and 30 s of plasma treatment resulted in about 2.6 and 3.6 log cycles reductions in Salmonella population (with the bacteria counts of 5.98 ± 0.09 CFU cm⁻² and 4.96 ± 0.11 CFU cm⁻², respectively). The greatest decontamination effect of plasma on eggshell was observed when the eggs were treated by plasma for 40 s and bacteria population was reduced from 8.58 ± 0.24 CFU cm⁻² to 4.49 ± 0.15 CFU cm⁻², i.e., 4.09 log cycle reductions in Salmonella population. The decontamination effect of cold plasma has been previously reported and discussed in previously published papers (Mandal et al. 2018). Also, the efficacy of cold plasma for decontamination of hen egg was reported in the literature (Apostol et al. 2015; Georgescu et al. 2017; Wan et al. 2017). For example, Wan et al. (2017) showed the effectiveness of a high voltage atmospheric cold plasma system against Salmonella inoculated on the hen eggshell (Wan et al. 2017). Similarly, Apostol et al. (2015) evaluated the bactericide effects of a high voltage pulsed jet plasma on the hen eggs that were contaminated with S. enterica using a mixture of helium and oxygen as the carrier gas at the voltage of 20–25 kV. They observed complete decontamination after 5 min of plasma treatment (Apostol et al. 2015). These results were in line with the observation in this study. However, a greater decontamination effect was reported by Apostol et al. (2015). This could be related to the use of different types of eggs (hen egg vs. duck egg), different types of plasma systems, and different plasma treatment conditions (e.g. treatment time). Regarding the type of sample, researchers previously showed that the more porous structure of duck eggshell (as compared to that of hen eggshell) can increase the possibility of bacterial survival during decontamination processes of eggs (Kaewmanee et al. 2013; Quan and Benjakul 2019). Besides, it was previously explained that the type and the concentrations of plasma-generated reactive species (e.g. charged particles, ozone, and UV radiation), which are believed to have decontamination effects, varies depending on the plasma-generating systems (Gavahian and Khaneghah 2019; Gavahian and Cullen 2019; Gavahian et al. 2019). Besides, Apostol et al. (2015) used a longer plasma treatment time that the present study (300 s vs. 40 s) which could be one of the reasons for the higher decontamination effects of plasma in their study. Similarly, in the present study, it was observed that increasing the plasma treatment time from 10 to 40 s enhanced the decontamination effects. In accordance to the findings of the present study, Wan et al. (2017) hypothesis that longer exposure of bacteria to plasma-generated reactive species can result in a better bactericidal effect of cold plasma (Wan et al. 2017). Similarly, Hertwig et al. (2017) reported that increasing the plasma treatment time from 2.5 to 15 min resulted in a greater reduction in the Salmonella population (3 log cycle reductions vs. 6 log cycle reductions) on the surface of almond samples (Hertwig et al. 2017). However, it should be noted that longer arc plasma treatment times can result in the denaturation of egg albumin protein due to the elevated temperatures (Figure S1). Moreover, the reduced process time is among the benefits of disruptive emerging technologies, such as cold plasma, which may provide higher profit for the industry (Gavahian et al. 2018a). Hence, prolonged process times can negatively affect the industrial attractiveness of this decontamination technique.

Fig. 3.

Effects of plasma treatment time on the Salmonella count (blue columns) and Salmonella reduction (orange line). Control: untreated egg (contaminated egg); 10 s: plasma-treated eggs for 10 s; 20 s: plasma-treated eggs for 20 s; 30 s: plasma-treated eggs for 30 s; 40 s: plasma-treated eggs for 40 s (color figure online)

Effects of plasma treatment on physicochemical properties of eggs

Eggshell strength, thickness and color

The results showed that plasma treatments did not affect the strength of eggshells (ranged from 48.2 N to 49.4). Similarly, all the plasma-treated samples had similar eggshell thickness to that of untreated sample (Table 1). Besides, none of the plasma treatments substantially changed the color of eggshells (ΔE ≤ 0.4). Cserhalmi, et al. (2006) classified the ΔE values as great (6.0–12.0), well visible (3.0–6.0), noticeable (1.5–3.0), slightly noticeable (0.5–1.5), and not noticeable (0–0.5) (Cserhalmi et al. 2006). Moreover, there was no significant difference between the ΔE values of the plasma-treated samples (Table 1). Similarly, Dasan et al. (2018) did not observe any significant color change in the plasma-treated hen eggs (Dasan et al. 2018). Puligundla et al. (2018) reported that color differences were only noticeable when the Gwamegi samples were treated by plasma for a long time (more than 6 min) (Puligundla et al. 2018). According to the authors, 3 min of plasma treatment slightly changed the Gwamegi color but 10 min of this treatment resulted in very distinct color changes. In the present study, the plasma treatment did not affect the eggshell color probably due to the short treatment time (treatment time < 0.7 min) and relatively low temperature of the process (surface temperature < 50 °C). Color differences are obvious to the human eye when ΔE is above 3.0 (Puligundla et al. 2018). As a result, a similar visual color was observed for all the studied duck egg samples. Therefore, unpleasant changes in the eggshell color probably is not a concern if a similar arc plasma process can be applied to duck egg in a continuous system. Future upscaling studies can further assess this aspect as the visual color of the eggshell is among the determining quality parameters of duck eggs (Gavahian et al. 2019).

Effects of plasma on chemical properties of egg

Effects on albumen and yolk pH, yolk color, yolk height, and Haugh unit

Table 2 presents the effects of plasma treatment on the pH of egg albumen and yolk. Forty seconds of plasma treatment slightly decreased the pH of the yolk from 6.20 to 6.04. The pH of egg albumens ranged from 8.79 to 8.59 and was similar to the previous reports on the pH of fresh eggs (Akter et al. 2014; Quan and Benjakul 2018). The results showed that there is no significant difference between the albumen pH of plasma-treated samples and untreated samples. It was previously explained the egg albumen contains a carbonate-bicarbonate buffer system and the changes in the albumen pH depends on its buffering capacity (Akter et al. 2014). It was reported that plasma treatment can drastically reduce the pH of water by acid formation through chemical reactions between water molecules and plasma-generated reactive species (Thirumdas et al. 2018). For example, Ma et al. (2015) reported that 10 min of plasma treatment decreased the pH of pure water from 7.0 to 3.2 (Ma et al. 2015). It was explained that the amount of pH variation depends on the treated material and plasma process conditions (e.g. carrier gas, treatment duration, and plasma equipment design) (Thirumdas et al. 2018). In the current study, the plasma was applied to the duck egg which has a different chemical composition and a limited amount of free H₂O molecules when it compares to the pure water. Furthermore, cold plasma has a low penetration depth and mainly affects the surface of the food materials (Ulbin-Figlewicz et al. 2013; Kim et al. 2015; Mandal et al. 2018; Xiang et al. 2018). In addition, plasma treatment was applied for a short time (< 0.7 min) in the present study. These could be possible reasons for the slight reduction in pH values of plasma-treated samples. Similarly, Wan et al. (2017) reported that the egg yolk pH was not affected during dielectric barrier discharge plasma treatment of hen eggs and was about 6.2. Likewise, Georgescu et al. (2017) treated hen egg with direct (carrier gas: a mixture of helium, oxygen and water vapor) and indirect (carrier gas: atmospheric air) dielectric barrier discharge. They observed that the albumen pH of the eggs changed slightly from 9.40 to 9.42 and 9.38 after direct and indirect plasma treatments, respectively (Georgescu et al. 2017). Besides, yolk colors of plasma-treated samples (ranged from 13.80 ± 0.45 to 14.00 ± 0.89) were similar to that of the control sample, i.e., 13.67 ± 0.58 (Table 2). Also, yolk height and Haugh unit of eggs were not affected by plasma treatments (Table 2). Similar results were observed when hen eggs were treated by cold plasma (Wan et al. 2017). Therefore, it seems that plasma treatment did not alter the studied physical parameters, i.e., Haugh unit, yolk color, yolk height, yolk pH, and albumen pH) of duck eggs.

Table 2.

Effect of plasma treatment on some internal quality parameters of eggs

	Albumen pH	Yolk pH	AV (mg KOH g−1)	TBARS (μgMD mL−1)	Yolk color	Yolk height (mm)	Haugh unit
Control	8.77 ± 0.05a	6.20 ± 0.02a	3.33 ± 0.01a	2.46 ± 0.25a	13.67 ± 0.58a	4.51 ± 0.58a	67.70 ± 3.92a
10 s	8.78 ± 0.16a	6.13 ± 0.03b	3.19 ± 0.05b	2.35 ± 0.11a	14.00 ± 0.89a	4.37 ± 0.61a	66.30 ± 6.19a
20 s	8.59 ± 0.21a	6.19 ± 0.07ab	3.36 ± 0.11a	2.44 ± 0.18a	13.80 ± 0.45a	4.40 ± 0.60a	65.60 ± 3.67a
30 s	8.66 ± 0.16a	6.10 ± 0.01a	3.34 ± 0.05a	2.40 ± 0.08a	14.00 ± 0.89a	4.57 ± 0.63a	66.86 ± 5.50a
40 s	8.79 ± 0.10a	6.04 ± 0.02c	3.23 ± 0.06ab	2.44 ± 0.16a	14.00 ± 0.71a	4.52 ± 0.77a	67.30 ± 4.20a

*Data are presented as Means ± standard deviation within a column with the same lowercase letters are not different significantly at p > 0.05. Control: untreated egg; 10 s: plasma-treated eggs for 10 s; 20 s: plasma-treated eggs for 20 s; 30 s: plasma-treated eggs for 30 s; 40 s: plasma-treated eggs for 40 s

Effects on acid value and TBARS value of yolk oil

The TBARS (ranged from 2.35 to 2.46 μg MDA mL⁻¹) and AV (ranged from 3.19 to 3.36 mg KOH g⁻¹) of egg yolks were not noticeably affected by plasma treatments (Table 2). Similar findings on the effects of plasma treatment on TBARS values were previously observed for other lipid-containing foods (Gavahian et al. 2018a). Likewise, Puligundla et al. (2018) reported that the AV of a semi-dried Pacific saury was not affected by corona discharge plasma treatment (Puligundla et al. 2018). Conversely, it was previously reported that plasma treatment can cause lipid oxidation in some food materials (Gavahian et al. 2018a). However, the plasma-induced lipid oxidation can be affected by plasma process parameters, including the treatment duration and the type of treated food (Gavahian et al. 2018a). Furthermore, it is generally believed that plasma has a low penetration depth (Gavahian and Cullen 2019). These are probably the reasons for the ineffectiveness of plasma treatment on internal parts of the duck eggs (e.g. yolk and albumen) in the present study.

Table 3 represents the fatty acid compositions of the egg yolk of the untreated and plasma-treated duck eggs. The fatty acid compositions of all the studied duck eggs were similar to those reported by Ali et al. (2017) and Ruan et al. (2018). The major fatty acid (> 48%) in all the studied yolk samples was C18:1. Similarly, Ruan et al. (2018) reported that the most abundant fatty acid in duck egg was the oleic acid with the concentration of 44–48%, depending on the farming conditions (e.g. the diet of ducks). It was previously explained that the fatty acid composition of the duck egg can be also affected by the age of ducks (Ali et al. 2017; Ruan et al. 2018). In the present study, the major fatty acid (oleic acid) was slightly decreased from 49.1 ± 0.4 to 48.6 ± 0.2% after 40 s of plasma treatment. All the changes in the percentage of fatty acids after plasma treatment were less than 0.5% and the ratio of unsaturated fatty acids to saturated fatty acid remained stable (with the average value of 2.1 and the maximum variation of 0.1%) after plasma treatments. It could be concluded that plasma treatment did not noticeably change the fatty acid compositions of the egg yolks. It seems that the plasma-generated reactive species were not in contact with the egg yolk as they need to pass both eggshell and egg white. The low penetration depth of plasma reactive species was previously discussed in the literature (Ulbin-Figlewicz et al. 2013; Kim et al. 2015; Mandal et al. 2018; Xiang et al. 2018). Therefore, it could be mentioned that arc plasma treatment did not considerably affect the internal parts of duck eggs (e.g. egg yolk). Concurrent with this result, TBARS values of the egg yolks were not changed significantly (p > 0.05) after 40 s of plasma treatment. Regardless of the raised concern about the plasma-induced lipid oxidation of the food, it seems that egg yolk could be safe from this possible negative impact while disinfecting the eggshell with plasma for a short time. Therefore, arc plasma treatment can be regarded as a potential tool for industrial decontamination of duck eggs.

Table 3.

The fatty acid composition of the extracted oil from duck egg yolk

Fatty acid	Relative concentration (%)
	Control	10 s	20 s	30 s	40 s
C14:0	0.5 ± 0.0a	0.5 ± 0.0a	0.5 ± 0.1a	0.5 ± 0.0a	0.5 ± 0.1a
C15:0	0.1 ± 0.0a	0.1 ± 0.0a	0.1 ± 0.0a	0.1 ± 0.0a	0.1 ± 0.0a
C16:0	23.6 ± 0.1a	23.5 ± 0.2a	23.3 ± 0.0b	23.4 ± 0.2a	23.1 ± 0.0b
C16:1	2.3 ± 0.1a	2.2 ± 0.2a	2.1 ± 0.1a	2.2 ± 0.0a	2.2 ± 0.1a
C17:0	0.4 ± 0.0a	0.3 ± 0.0b	0.3 ± 0.0b	0.3 ± 0.0b	0.2 ± 0.2ab
C17:1	0.3 ± 0.0a	0.3 ± 0.0a	0.3 ± 0.0a	0.3 ± 0.0a	0.3 ± 0.0a
C18:0	5.5 ± 0.0a	5.5 ± 0.3a	5.9 ± 0.4a	5.7 ± 0.4a	5.4 ± 0.4a
C18:1	49.1 ± 0.4a	49.0 ± 0.3a	48.6 ± 0.7a	48.5 ± 0.4a	48.6 ± 0.2a
C18:2	11.3 ± 0.2a	11.3 ± 0.4a	11.1 ± 0.0a	11.1 ± 0.3a	11.6 ± 0.1a
C18:3	1.3 ± 0.0a	1.1 ± 0.1b	1.2 ± 0.1ab	1.3 ± 0.0a	1.2 ± 0.1ab
C20:1	0.6 ± 0.1a	0.6 ± 0.1a	0.6 ± 0.1a	0.7 ± 0.1a	0.7 ± 0.1a
C20:2	0.4 ± 0.0b	0.4 ± 0.0b	0.4 ± 0.0b	0.4 ± 0.0b	0.5 ± 0.0a
C20:3	0.3 ± 0.0a	0.3 ± 0.0a	0.3 ± 0.0a	0.3 ± 0.0a	0.3 ± 0.0a
C21:0	0.1 ± 0.0a	0.1 ± 0.0a	0.1 ± 0.0a	0.1 ± 0.0a	0.1 ± 0.0a
C22:0	0.1 ± 0.0b	0.1 ± 0.0b	0.1 ± 0.0b	0.1 ± 0.0b	0.2 ± 0.0a
C23:0	2.0 ± 0.1c	2.2 ± 0.1abc	2.3 ± 0.0a	2.2 ± 0.0b	2.3 ± 0.1ab
C20:5	0.1 ± 0.0a	0.1 ± 0.0a	0.1 ± 0.0a	0.1 ± 0.0a	0.1 ± 0.0a
C22:6	2.1 ± 0.3a	2.3 ± 0.2a	2.5 ± 0.6a	2.5 ± 0.6a	2.5 ± 0.5a
SFA	32.3 ± 0.1a	32.3 ± 0.6a	32.6 ± 0.3a	32.4 ± 0.6a	31.9 ± 0.6a
UFA	67.7 ± 0.1a	67.7 ± 0.6a	67.4 ± 0.3a	67.6 ± 0.6a	68.1 ± 0.6a
UFA/SFA	2.1 ± 0.0a	2.1 ± 0.1a	2.1 ± 0.0a	2.1 ± 0.1a	2.1 ± 0.1a

*Only fatty acids with relative concentrations of above 1% are reported. Data are presented as Means ± standard deviations within a row with the same lowercase letters are not different significantly at p > 0.05. UFA: total fatty acid content; SFA total saturated fatty acid content; UFA/SFA the ratio of unsaturated fatty acids to saturated fatty acids. The fatty acids are presented as C:D which represents number of carbons and double bonds, respectively. Control: untreated egg; 10 s: plasma-treated eggs for 10 s; 20 s: plasma-treated eggs for 20 s; 30 s: plasma-treated eggs for 30 s; 40 s: plasma-treated eggs for 40 s

Conclusion

This study, for the first time, revealed the applicability of arc plasma for duck egg decontamination. The results showed that process optimization, in terms of exposure time, can minimize the bacteria count of eggshells and prevent the unpleasant changes (e.g. albumen denaturation) of the plasma-treated duck eggs. Also, the eggshell thickness, strength, and color were not affected by plasma treatments due to the short exposure time and the low temperature of the process. Similarly, the internal quality (Haugh unit, yolk color, yolk height, and pH) of plasma-treated samples were similar to those of untreated eggs. Besides, the physicochemical properties of yolk (e.g. fatty acid composition, AV, and TBARS) were not affected by plasma treatment probably due to the low penetration depth of plasma-generated reactive species. However, further non-thermal reduction of Salmonella population through the arc plasma was not feasible in the present study due to the elevated temperatures at the extended treatment times. Therefore, prospective studies may explore the applicability of various plasma sources with improved designs for the production of Salmonella-free duck eggs. Moreover, up-scaling and economic studies are among the prerequisites for the successful commercial application of this emerging technique.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Compliance with ethical standards

Conflict of interest

The authors have no conflict of interest to declare.