The effect of atmospheric pressure cold plasma on the inactivation of Escherichia coli in sour cherry juice and its qualitative properties

Abstract One of the nonthermal methods is the atmospheric pressure cold plasma (APCP). In this study, the effect of cold plasma on the reduction of Escherichia coli bacteria and qualitative properties of sour cherry juice, including total phenolic content (TPC), total anthocyanin content (TAC), and vitamin C, were investigated. Independent variables included plasma exposure time (1, 5, and 9 min), applied field intensity (25, 37.5, and 50 kV/cm), feeding gas oxygen content (0%, 0.5%, and 1%), and sample depth (0.5, 1, and 1.5 cm). The results show that increased oxygen content in argon has the greatest effect on the reduction of bacteria, and plasma exposure decreased 6 logarithmic periods of E. coli bacteria in sour cherry juice. Optimization results showed when all bacteria were eliminated by plasma, TPC remained unchanged, and TAC and vitamin C decreased by 4% and 21%, respectively, while thermal methods increased TPC by 23% and decreased TAC and vitamin C by 26% and 77%, respectively. These results indicate that, compared with conventional thermal methods, sour cherry juice pasteurization using APCP has little effect on the juice qualitative properties, and this method can serve as a suitable alternative to conventional thermal methods.

In this regard, researchers studied the reduction of E. coli in raw milk by using cold plasma (Coutinho et al., 2019;Gurol, Ekinci, Aslan, & Korachi, 2012). In another study, the effects of cold plasma on the inactivation of three types of bacteria including Staphylococcus aureus, E. coli, and Candida albicans in orange juice and its qualitative properties were investigated (Shi et al., 2011). Investigations have shown that the use of the APCP in reducing the microbial load of liquid food has been in cases that the low volumes and low depths of liquids were tested (Gurol et al., 2012;Shi et al., 2011;Surowsky, Fröhling, Gottschalk, Schlüter, & Knorr, 2014) and research on the inactivation of bacteria in greater depths of liquids seems necessary.
Since the sour cherry with Prunus cerasus L. scientific name is one of the most important horticultural products across the world, about 85 percent of this product is used in a processed form such as juice, jam, and etc. (Toydemir et al., 2013). The aim of this study was to investigate the effects of the DBD-APCP on the inactivation of E. coli in sour cherry juice and its qualitative properties. Then, the results were compared with the results of the conventional thermal method.

| Generation of cold plasma
In this study, dielectric barrier discharge (DBD) was used to generate plasma. The geometry of the plasma generation system is very flexible in the DBD method (Misra, Schlüter, & Cullen, 2016). In this study, coaxial electrodes were used to create a plasma generation system.
The fabricated nozzle includes a ceramic tube dielectric, a copper ring electrode, and a tungsten central electrode (Figure 1), and these electrodes were connected to an AC power supply. The output voltage of the power supply varied between 0 and 20 kV, and the frequency was set at 20 kHz. The diameter of the central electrode 1 mm, the gap between the central electrode and the dielectric 4 mm, the internal radius and length of the ring electrode 5.5 and 10 mm, respectively, the thickness, internal radius, and length of the dielectric are 1 mm, 4.5 mm, and 10 cm, respectively, and the feeding gas flow 5 slm (standard liter per minute) were considered (Gas flow was measured and controlled by a flowmeter), and the combination of argon and oxygen was used as feeding gas (Surowsky et al., 2014). Samples were placed in circular glass plates of 2 cm in diameter. The distance between the surface of the samples and the nozzle was adjustable and was set at 2 cm for all experiments. Figure 1 shows the schematic illustration of the plasma generation system designed in this study (Video S1). Plasma properties were determined by OES and temperature measurement. In order to identify the plasma species, the emitted spectra were recorded by using a spectrophotometer (Ocean Optic, HR2000+CG) in the wavelength range of 200-900 nm (Figure 1). For this purpose, the detector was placed perpendicularly to the nozzle axis at a distance of 2 cm from the nozzle. Then, the data obtained through the spectroscopy were analyzed by using the "AvaSoft" software version 8 and the related diagrams are plotted, and the peak locations were compared with the NIST atomic spectra database to identify the particles generated from the plasma (Huang, Yu, Hsieh, & Duan, 2007;Surowsky et al., 2014).
Plasma coldness was also determined by measuring the temperature of the samples. Temperature measurements were performed for all samples before and after plasma exposure (Surowsky et al., 2014).

| Preparation of microbial samples and microbial tests
For this purpose, a certain amount of sour cherry fruit of Mashhad Champa variety was purchased from local markets. First, the fruits were washed, cleaned, and cored. The prepared fruits were then dewatered using an electric juicer. In order to separate pulp suspensions and tissue components, the extracted juice was poured into a centrifuge with the speed of 6,000 rpm (4,307 g) for 20 min. For complete separation of the remaining suspended particles, the transparent portion of the extract was passed through a Whatman filter paper using a vacuum pump (Hosseinzadeh Samani, Minaee, & Khoshtaghaza, 2015).
The samples were then sterilized in autoclaved glass containers and sterilized using an autoclave at 121°C, under 15 psi for 15 min. Next, the sour cherry juice samples were contaminated with E. coli bacteria to determine the reduction rate of bacteria in each sample after undergoing the process of interest. To do this, a few colonies of E. coli bacteria cultured 24 hr ago were transferred into a sterile test tube containing physiological serum (sodium chloride 9%). Then, the samples were contaminated by adding an appropriate volume of a solution prepared with autoclaved sour cherry juice (1:9 v/v). The bacteria was counted before and after the juice samples were treated by plasma. To count bacteria, standard (viable) plate count method was used, and for each sample, five dilutions were obtained, and from each dilution, 1 ml was cultured on MacConkey Agar culture medium (all of these steps were carried out under a hood). The prepared Petri dishes were then incubated for 24 hr at 37°C to encourage the growth of E. coli bacteria. Finally, the number of colonies per plate was determined and the bacteria in each sample were counted (Deng, Cheng, Ni, Meng, & Chen, 2010).

| Evaluation of chemical and qualitative properties of sour cherry juice
In the next step, the chemical and qualitative properties of the sour cherry juice were evaluated by measuring the total phenolic content (TPC), total anthocyanin content (TAC), and vitamin C. It should be noted that the unsterilized sample was used for plasma treatment and chemical evaluation.

| Measurement of TPC
Total phenolic content was measured by using the Folin-Ciocalteu colorimetric method. For this purpose, a volume equivalent to 0.0025 g of the sour cherry juice was mixed with 1 ml of methanol. The volume of the resulting mixture was increased to 2.25 ml by addition of distilled water. Then, 0.25 ml of Folin-Ciocalteu (10% v/v) was added to the mixture and shaken to mix thoroughly. After 5 min, 2 ml of sodium carbonate "Na 2 CO 3 (20% v/v)" was added. The samples were placed in the dark at 25°C for 2 hr, and then, the absorbance was read at a wavelength of 765 nm by using a spectrophotometer (UV-1800). By placing the results obtained from the absorbance of the samples in the calibration curve, the TPCs of the samples were measured and expressed as mg of gallic acid equivalent (GAE) in 100 g of the sour cherry juice. All experiments were performed in triplicate. To plot the standard curve, various dilutions of gallic acid were prepared and the standard curve is drawn (Liang, Yue, & Li, 2010).

| Measurement of TAC
The TACs of samples were measured by using the pH differential method and based on cyanidin-3-glucoside as dominant anthocyanin of sour cherry (Blando, Gerardi, & Nicoletti, 2004). To this end, two buffers, "potassium chloride (KCL, pH = 1)" and "sodium acetate (HCL, pH = 4.5)" were used. For each experiment, 2 ml of each sample was poured into two 25-ml volumetric flask, and then, one of the balloons was reached to volume by the KCL buffer and the other one was reached to volume by adding HCL buffer. After 20-40 min, at wavelengths of 520 and 700 nm, the absorbance was read by the UV-1800 spectrophotometer two times, once with the KCL buffer and again with the HCL buffer, and the TAC was measured by Equation 1. where Mw is molar mass of cyanidin-3-glucosides, ε: molar extinction coefficient for cyaniding-3-glucoside, DF: dilution factor and L: cuvette optical path length, and A: absorbance, which was calculated by where A 520 is absorbance at 520 nm and A 700 : absorbance at 700 nm wavelength (Lee, Durst, & Wrolstad, 2005;Rouhani, Eyenafshar, & Ahmadzadeh, 2015). All experiments were carried out in triplicate.

| Determination of vitamin C
Concentration of vitamin C was determined used by a redox titra- (1) (3) Ascorbic acid + I 2 → 2 I − + dehydroascorbic acid F I G U R E 1 The schematic illustration of the plasma generation system In Equation 3, I − is iodide ions and I 2 is iodine (Anonymous, 2018).

| Thermal pasteurization methods
First, the 5 ml of sour cherry juice samples were pasteurized by a conventional thermal method at a temperature range of 85°C for 5 min . Then, TPC, TAC, and vitamin C level were determined and the results were compared with the results obtained through the APCP method.

| Analysis, modeling, and optimization of experiments
In this study, the effects of DBD-APCP on the inactivation of E. coli in the sour cherry juice and its qualitative properties were investi- In this study, independent variables were plasma exposure time (min), the depth of samples (cm), the average field strength created at the distance between the electrons (in this article referred to as field strength) (kV/cm), and the oxygen content in argon (%), whose effects of them on the inactivation of E. coli bacteria, and the levels of TPC, TAC, and vitamin C in sour cherry juice were studied.
The range of changes in independent variables in the experiments is shown in Table 1 (encoded at three levels). Under these conditions, 30 different tests were designed.

| Characteristics of the plasma source
The curves of the emission spectra, for pure argon, 99.5% argon and 0.5% oxygen, and 99% argon and 1% oxygen, are shown in Figure 2 The intensity of the lines corresponding to the argon atoms in the range of 650-900 nm is high in all curves (Huang et al., 2007;Mohamed, Al Shariff, Ouf, & Benghanem, 2016). In this range, the lines corresponding to the oxygen atoms at 777.4 and 844.5 nm wavelengths are also observed. Also, the reaction of free electrons with N 2 and O 2 molecules results in the formation of excited N 2 molecules, N + 2 , N + , O 3 , and a little of NOx in the UV-C range. Due to the reaction of reactive species obtained from the plasma with water in the air and sour cherry juice, OH and O radicals are produced (Huang et al., 2007;Mohamed et al., 2016). Figure 2 shows the concentrations of O, OH, and O 3 radicals, when feeding gas contains 0.5% and 1% oxygen, compared to the state that feeding gas is pure argon, will increase by 205%, 38%, 16%, and 359%, 83%, and 34%, respectively.
In this study, the nonthermally of the generated plasma was determined by measuring the temperature of the samples. Before the tests, the temperature of all samples was about 21°C, and after plasma exposure, the temperature varied from 17 to 21°C depending on the test conditions. By increasing the plasma exposure time and reducing the depth, the temperature of the samples was further reduced. Reducing the temperature at higher field strengths is less due to the higher temperature of the plasma jet. Figure 3 shows the changes in temperature and pH in the samples after plasma exposure. The reduction of temperature is due to the lower temperature of the feeding gas than the samples (Surowsky et al., 2014).

| The effect of plasma on the inactivation of E. coli bacteria
The ANOVA results show the effects of plasma exposure time, depth of samples, field strength, and oxygen content in argon, as well as the effects of time × field strength, time × oxygen percentage in argon, depth of samples × oxygen percentage in argon and field strength × oxygen percentage on the inactivation of the E. coli bacteria in sour cherry juice are significant (p < .01; Table 2). The total sum of squares obtained from ANOVA shows that, among the main variables, the percentage of oxygen in argon accounts for 67.5% of variance in data has the greatest effect on the inactivation of bacteria, followed by field strength, plasma exposure time, and depth of sample, explaining 12%, 6.8%, and 4.3%, of variance in the data, respectively.    Table 2 also shows, represents the greater effect of these two variables on bacterial inactivation compared to other variables. In Equation   5, the negative coefficient of selected variables indicates that with increasing each of them, the number of E. coli bacteria decreases.  3 min of plasma exposure (Gurol et al., 2012). Increasing exposure time from 1 to 8 min reduced 5 logarithmic cycles of Citrobacter freundii bacteria in apple juice (Surowsky et al., 2014).
As well as, according to Figure  Because the structure of the liquid foods is different from the solids and different layers of the liquids can be exposed to the plasma, and the liquids have water molecules that can react with ROS, the depth at which plasma can inactivate the bacteria is greater than the corresponding depth in solid materials (Misra et al., 2016).
In the study, as shown in Figure  The field strength applied in the region of generation plasma is directly correlated with the applied voltage (Misra et al., 2016).
Therefore, in our study, changing the voltage of the power supply was used to change the field strength. Figure 5a, This reduction can be attributed to increasing in feeding energy due to increased field strength and, consequently, increased the concentration of ROS derived from plasma (Lu, Patil, Keener, Cullen, & Bourke, 2014;Misra et al., 2016). Figure 5a shows that the effect of increasing field strength on the inactivation of bacteria in longer exposure times (over 5 min) is greater than that in shorter (<5 min) ones (Bermúdez-Aguirre et al., 2013), because in this condition, the concentrations of plasma species at a certain interval are more than those when field strength is lower (Noriega et al., 2011).
In this study, the pure argon and the combination of argon and oxygen were used for plasma generation. The OES results ( Figure 2) and ANOVA results (Table 2) and Figure 5c,e,f show that by adding

F I G U R E 4
Verification of the actual data versus the data obtained from the model Also, the obtained results from Table 2 and Figure 5f showed that the effect of increasing field strength on the inactivation of E. coli bacteria was more pronounced when the feeding gas contained a higher percentage of oxygen, because simultaneous increases in field strength and oxygen content produce high concentrations of ROS (Huang et al., 2007;Perni, Liu, Shama, & Kong, 2008). Therefore, in this study, the greatest reduction in E. coli bacteria occurred when plasma exposure time, the field strength, and oxygen percentage in argon were high (9 min, 50 kV/cm and 1%, respectively), and the depth of the sour cherry juice samples was minimum (0.5 cm). This value is equal to 6 logarithms cycles. the experiment's conditions. In order to study this, the pH of the samples before and after the experiments was measured, and it was observed that the pH of the untreated samples was 2.99, and after the plasma exposure, the changes in sour cherry juice sample's pH, according to the experiment's conditions, were obtained 0.03 ( Figure 3). The small effect of this change does not contribute greatly to reducing the sour cherry juice microbial load (Shi et al., 2011). UV photons generated by the APCP can also be effective in destroying bacteria. The results of the OES showed that UV emission in this study (at wavelengths below 280 nm) was negligible. In addition, the researchers have argued the UV produced is rapidly weakened and does not reach the liquid surface. Therefore, UV emissions have little effect on the inactivation of E. coli bacteria in sour cherry juice (Deng, Shi, & Kong, 2006;Laroussi & Leipold, 2004;Shi et al., 2011).

| The mechanism of inactivation of E. coli in sour cherry juice by APCP
According to studies on the destruction of bacteria by cold plasma, it can be argued that charged particles (e.g., positive and negative electrons) and ROS play the most important role in inactivating bacteria (Deng et al., 2010;Li et al., 2013;Mohamed et al., 2016). The reactions that cause the formation of ROS are shown in Equations 7-11 (Bruggeman et al., 2016;Misra et al., 2016;Shi et al., 2011): Researchers have argued that ROSs created by APCP and charged particles can cause damage to bacteria by morphological changes in cell and change in the structure of cell membrane (Ali et al., 2014;Deng et al., 2006;Yang et al., 2011). The cell walls in gram-negative bacteria such as E. coli are covered by a thin layer of peptidoglycan and an outer membrane of lipopolysaccharide.

TA B L E 3 The analysis of variance results for the effect of ACPC on qualitative properties (TPC, TAC and vitamin C) in sour cherry juice
During plasma exposure, ROS can react with peptidoglycan and lipopolysaccharide and, by breaking down the C-O, C-N, and C-C bonds, damage the molecular structure of these materials and destroyed them (Han et al., 2016). It should be noted that increased plasma exposure time causes more bacteria to be damaged and, in many cases, many deformities occur in their bacterial cell membrane and appearance (Ali et al., 2014;Han et al., 2016;Yang et al., 2011).

| The effect of plasma on the qualitative properties of sour cherry juice
The ANOVA results regarding the effect of APCP on the TPC, TAC, and vitamin C are shown in Table 3, and the obtained models for TAC in the encoded and actual conditions are presented in Equations 12 and 13 and for vitamin C in Equations 14 and 15.
The explanation coefficient and standard error of the models were 0.967 and 0.70 for TAC and 0.972 and 0.51 for vitamin C, respectively. Table 3 shows the effects of plasma exposure time, depth of samples, field intensity, and oxygen content in argon on the TPC of the sour cherry juice samples were not significant (p > .01) but those on the TAC and vitamin C of the juice were significant (p < .01).
The TPC of the untreated sour cherry juice samples was calculated at 277.05 ± 2.1 mg GAE/100 g and that of the treated samples ranged from 273.01 ± 1.07 to 276.28 ± 0.91 mg GAE/100 g. These results indicated that the TPC in the sour cherry juice remained unchanged after plasma exposure (Table 3). This result is consistent with the results of a study that investigated the effect of cold plasma on the chemical properties of strawberries (Misra, Pankaj, Frias, Keener, & Cullen, 2015).
However, researches' results in this regard are contradictory. The results of some studies indicate that plasma exposure increases the TPC in pomegranate juice  and sour cherry juice (Garofulić et al., 2015). In addition, plasma exposure reduced the phenolic compounds of the lettuce leaf samples (Grzegorzewski, Ehlbeck, Schlüter, Kroh, & Rohn, 2011), but we observed no change in the TPC of the sour cherry juice due to plasma exposure in the current study (p > .01). Table 3 shows that plasma exposure alters the TAC of the sour cherry juice. The TAC of untreated samples, and the lowest and high- Confirmation of this is that by increasing ROS-enhancing agents, the anthocyanin in the sour cherry juice was a further reduction ( Figure 6a,b).
The results of studies on the effect of plasma on the TAC are also inconsistent. The study of Lacombe et al. (2015) illustrates that being prone to plasma for 90 s lead to a significant reduction in the anthocyanin content of the blueberry samples (Lacombe et al., 2015).
Likewise, the research has been carried out by  and reveals ozone could reduce more than 90% of the anthocyanin present in the blackberry juice. However, some studies have shown plasma exposure increases the anthocyanin content of the sour cherry juice by 15% (Garofulić et al., 2015) and increases the anthocyanin content of pomegranate juice . Misra et al. (2015) reported plasma had no effect on the anthocyanin content of strawberries.
Vitamin C of untreated samples was calculated at 36.03 ± 1.24 mg/L, but that of plasma-treated samples varied from 27.92 ± 0.9 to 5.79 ± 2.20 mg/L. The ANOVA results (Table 3) show plasma has a significant effect on vitamin C in the sour cherry juice samples (p < .01). According to Equation 14 and Table 3, it can be inferred that increasing the oxygen content in argon, in comparison with other variables, explains a higher percentage of variance in vitamin C of the samples. Therefore, the presence of molecules and oxygen radicals can be considered as the main cause of the reduction in vitamin C of the sour cherry juice. These particles lead to oxidation of vitamin C (Misra et al., 2015;Shi et al., 2011;Wang et al., 2012). The results of our study show that increasing oxygen content, increasing field intensity (Figure 7a), increasing plasma exposure time, and decreasing the depth of samples (Figure 7a), reduce the amount of vitamin C. The increase in plasma exposure time and applied voltage (directly correlated with field intensity) has been reported to reduce the vitamin C of the strawberry samples (Misra et al., 2015). Another study reported that plasma reduced vitamin C of pieces of cucumber, hawthorn, and pear by 3.6, 3.2, and 2.8, respectively (Wang et al., 2012). Also, in another study, ozone reduced vitamin C of strawberries by 50% within approximately 5 min (Tiwari, O'Donnell, Patras, et al., 2009). While, exposing DBD-plasma generated by air to orange juice for 5, 10, 15, and 20 s showed that plasma has no significant effect on vitamin C in the samples (p > .05 ;Shi et al., 2011).

| Optimizing the results and comparing with the results of the conventional thermal method
In this section, the optimal conditions for use of APCP were deter-  Table 4. The results of the optimization in two states are shown in Table 5.
The results of optimization show that under the boundary conditions (Table 4), the optimum exposure time, field intensity, oxygen percentage, and depth of samples were 8.8 min, 49 kV/cm, 1%, and 0.65 cm in two statuses 1 and 1 min, 50 kV/cm, 0.5% and 1.1 cm in two statuses 2, respectively. Table 5 shows the variation of qualitative properties of the sour cherry juice in states 1 and 2 as well. In this Table, the effects of thermal methods on the sour cherry juice also shown. As Table 5 shows, under optimal conditions in the state 1, plasma does not change in the TPC of samples and reduces their TAC and vitamin C content by 4% and 21%, respectively Misra et al., 2015;Wang et al., 2012). The decomposition of anthocyanins and ascorbic acid can be due to oxidative reactions caused by free radicals that are appeared in the plasma generation process. The greater decomposition of ascorbic acid, in comparison with anthocyanins, is due to its physical and chemical characteristics, because ascorbic acid is much more vulnerable against free radicals (Hosseinzadeh Samani et al., 2015).
Table 5 also shows that in state 2, with maximum preservation of the qualitative properties of sour cherry juice, 38% of E. coli bacteria in the sour cherry juice were destroyed. In this state, there was no change in the TAC and TPC, but the vitamin C decreased by 5%.
While in conventional thermal methods, the TPC of the sour cherry juice has increased due to exposure to high temperatures (above 80°C; Garofulić et al., 2015;Herceg et al., 2016). In this study, the increase in TPC in the thermal method is 26%. Studies have shown that phenolic compounds are formed at temperatures above 80-90°C, due to the supply of phenolic compounds' precursors as well as nonenzymatic exchanges between the molecules. In fact, in this temperature due to the inactivation of enzymes that are effective on decomposition and degradation, new phenolic compounds are formed. Also, pasteurization by conventional thermal method reduced the TAC and vitamin C content by 23% and 77%, respectively.
The reduction of vitamin C and TAC by the thermal method has already been confirmed (Samani et al., 2016). These results show that in comparison with the thermal methods, the pasteurization of the sour cherry juice by the APCP can help to maintain its qualitative properties. This result is consistent with the results of studies that have investigated plasma's effects on the qualitative properties of the products (Gurol et al., 2012; Note: Escherichia coli is expressed in terms of (CFU/ml), TAC in terms of (mg C3GE/L), TPC in terms of mg GAE/100 g, and vitamin C in terms of (mg/L). Different letters showed significant differences at 5% Duncan test.

| CON CLUS ION
In this study, the effects of the APCP on inactivation of E. coli bacteria in sour cherry juice and the qualitative properties of the juice were investigated. The results of this study indicate that by increasing the plasma exposure time, the field intensity, and the percentage of oxygen, and reducing the depth of samples, more E. coli bacteria were destroyed, and among these variables, increase of oxygen content plays the most substantial role due to the production of more ROS. The results also showed plasma had no effect on the TPC of the samples and, compared to the thermal methods, had a small effect on the TAC and ascorbic acid in sour cherry juice. This method can, therefore, be safely used to eliminate the bacteria without influencing the qualitative properties of the materials.

ACK N OWLED G M ENT
The authors are thankfully acknowledged Research Council of Shahrekord University for the financial support they gave to carry out the study.

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L S TATEM ENT
This study does not involve any human or animal testing.