Purpose

Fresh-cut minimally-processed vegetables are a rapidly developing class of convenience (ready-to-eat) foods and an important component of a healthy and balanced diet. However, major challenges facing the fresh produce industry include their rapid deterioration and limited shelf-life. Furthermore, green leafy vegetables such as spinach are typically linked to foodborne outbreaks, with irrigation/washing steps being acknowledged as major sources of microbial contamination.

The potential of cold plasma (CP) for mitigation of microbiological risks across food production systems has been demonstrated in the last decade. As CP treatment is limited to surface applications, which may negatively affect product quality and functionality, a recent application of this technology to overcome such limitations relies on the activation of liquids through their exposure to CP discharges, resulting in a cocktail of reactive oxygen and nitrogen species (RONS) causing the acidification of the media and an increase in the oxidation and reduction potential (ORP). PAW, with a series of advantages over direct CP treatment (e.g., dose control, storage capacity, offsite generation, self-sanitation), has recently gained growing interest as a sustainable, cost-effective alternative to current disinfectants. Nevertheless, further studies on PAW composition, stability and preservation ability on representative products, are still required towards technology upscaling and industrial uptake.

Main conclusions

 A remarkable dependency between PAW composition and CP operating parameters has been demonstrated:

• Increasing plasma power and activation time led to a significant drop in pH (2.4) and higher nitrates and nitrites levels (320 and 7.2 mg/L, respectively) and oxidation reduction potential (292 mV) in the PAW.
• Albeit it is acknowledged as one of the persistent species in the PAW commonly associated to its antimicrobial activity, hydrogen peroxide was not detected in PAW, irrespective of the treatment conditions, which was attributed to its instability in acidic environments and the remote PAW generation mode.

At least 2 weeks of chilled (4 °C) storage capacity (pH, ORP, nitrates, nitrites) have been confirmed, which will confer a more flexible operational margin to food producers to cope with the volatile supply and demand.

The potential of PAW for microbial inactivation, quality retention (colour) and shelf-life extension has been demonstrated for baby spinach leaves after 8 days of (air) storage at 4 °C:

• Rinsing steps influenced colour development during chilled storage to a greater extent than PAW treatment itself.
• About 1 log reduction in total bacterial counts was achieved through PAW rinsing, with no variability after 8 days at 4 °C (typical shelf-life at retailers). Moreover, microbial levels on PAW-treated samples after storage were significantly lower than those on control samples. The bactericidal activity is mainly attributed to RNS (NO₃, NO₂), low pH and high ORP.
• According to the Regulation (EC) 2073/2005, defining mandatory microbiological criteria for foodstuff, the bacterial flora normally present in ready-to-eat vegetables is likely to reach 6-7 log CFU/g. Likewise, evident organoleptic alterations occur in vegetables when total bacterial counts reach 7-8 log CFU/g. In the present study, microbial levels on PAW-treated samples after 8 days of storage at 4 °C reached about 5 log CFU/g, which leaves a relative margin for shelf-life extension beyond current cut-offs at retailers, thus contributing to reduced industry and household food waste generation.

Thus, PAW stands as a promising alternative to traditional sanitisers applied in the fresh produce industry. 

Methods

The composition of PAW generated from distilled water with a surface dielectric barrier discharge (SDBD) set-up was determined as a function of the CP power and activation time:

• A full factorial design was implemented (3 levels × 3 levels; at least in triplicate on independent days) to assess the effect of the plasma power (16 W, 26 W and 36 W) and activation time (5, 12.5 and 20 min).
• Spectroquant® test kits were used for spectrophotometric determination of nitrates and nitrites at 340 and 525 nm, respectively. Hydrogen peroxide was determined with the titanium sulphate colorimetric method at 407 nm.
• pH, ORP and immediate temperature reading were also determined in PAW samples.

PAW stability (RONS, pH, ORP) after 24 h, 1 wk and 2 wk of storage at 4 °C was determined (in triplicate) for all values of plasma power and 20 min CP activation time, since such an activation period resulted in the lowest pH, highest RONS and ORP, thus representing the most promising conditions for further food decontamination trials and industrial implementation.

Fresh-cut and unwashed baby spinach leaves were rinsed (in triplicate) with either PAW (36 W for 20 min) or distilled water (control) at 4 °C, for 2 min at 120 rpm on a laboratory rocker. The treatment was subsequently repeated to mimic industrial settings and the samples were then centrifuged for 5 min. Microbiological and colour analysis were conducted on both rinsed and untreated samples. Moreover, PAW-rinsed, control and untreated samples (three replicates) were packed in air and stored at 4 °C. The headspace was measured in three independent replicates with a Checkmate 9900 analyzer. Microbiological and colour analysis were conducted after 8 days of chilled storage (typical product shelf-life at retailers).

What was examined

PAW – pH

• The pH of PAW decreased as the plasma power and the activation time increased (pH 2.4 ± 0.1 for 36 W and 20 min). At constant plasma power, the pH drop was statistically more pronounced at shorter activation times (between 5 and 12.5 min), as no significant differences were found between 12.5 and 20 min exposure. At constant activation time, the acidification of the media was statistically similar for plasma power of 16 and 26 W, with more pronounced pH drop at 36 W.
• When the water was initially tempered at 4 and 10 °C, a significant increase in the pH (≈3.6) of the PAW (36 W for 20 min) was observed. 
• When the initial water volume was increased from 100 to 500 mL (44.8 vs. 32.0 mm gap distance between the electrode and the liquid surface), a significant increase in the pH (≈7.0) of the PAW (36 W for 20 min) was observed.
• When the initial pH of the water was adjusted to alkaline values with sodium pyrophosphate buffer 1% w/v (pH 9.7), significant differences were found in the pH (≈9.0) of the PAW (36 W for 20 min).

PAW – Temperature

• The temperature of PAW samples (immediate reading after CP treatment) increased towards CP activation time and plasma power, due to a thermal effect attributed to the SDBD set-up. A maximum temperature of 30.9 ± 1.2 °C (average initial temperature of 19.0 ± 0.7 °C) was reached for the most severe conditions of plasma power and activation time, and no variations in the volume of water before and after activation were recorded.
• When the water was initially tempered at 4 and 10 °C, a slight decrease in the temperature (≈28.0 °C) of the PAW (36 W for 20 min) was observed.  
• When the initial water volume was increased from 100 to 500 mL, a significant decrease in the temperature (≈22.0 °C) of the PAW (36 W for 20 min) was observed.

PAW – ORP (oxidation-reduction potential)

• PAW ORP values increased with higher plasma power and activation time, with values ranging between 200 and 292 mV for extreme operating conditions (16 W/5 min and 36 W/20 min).

PAW – Hydrogen Peroxide

• Hydrogen peroxide has not been detected in the PAW samples (no significant differences with respect to the blank), irrespective of the treatment conditions, albeit it is acknowledged as one of the persistent species in the PAW commonly associated to its antimicrobial activity.
• The absence of this compound in PAW has been attributed to the rapid decomposition of hydrogen peroxide by nitrites under acidic conditions. Moreover, the remote PAW generation mode (i.e., electrode not immersed in the water) and the relatively long activation times in the present study may have also affected its availability.

PAW – Nitrates and Nitrites

• Nitrates levels increased significantly towards both activation time and plasma power, with 36 W and 20 min yielding the highest concentration (320.0 ± 47.8 mg/L). Nitrites levels followed a similar evolution, although the maximum concentration achieved in PAW samples (7.2 ± 3.8 mg/L) was significantly lower, which has been attributed to their instability under acidic conditions, eventually decomposing into nitrates and nitrogen oxide. Nitrites levels for plasma power of 16 and 26 W were practically negligible at any activation time. Despite the relative variability among replicates, nitrites concentration at 36 W significantly increased after 12.5 and 20 min treatment.
• The energy density, i.e., the energy consumed for the creation of 1 mol of RNS, ranged between 7.5 and 8.1 × 10⁴ kJ/mol for extreme operating conditions (16 W/5 min and 36 W/20 min, respectively).
• When the initial water volume was increased from 100 to 500 mL, a significant decrease in the nitrates levels (≈40.0 mg/L) in the PAW (36 W for 20 min) was observed, although nitrites levels remained unaffected, most likely due to the neutral pH.
• When the initial pH of the water was adjusted to alkaline conditions (pH 9.7), no significant differences were found in nitrates and nitrites levels in the PAW (36 W for 20 min).
• When the water was initially tempered at 4 and 10 °C, nitrates levels in the PAW (36 W for 20 min) remained unaffected, although slight differences were observed in nitrites levels (≈28 mg/L).

PAW – Storage stability at 4°C

• For all the conditions assayed (16, 26, and 36 W for 20 min), pH, ORP, nitrates and nitrites levels in the PAW remained stable after 2 weeks of chilled storage (no statistically significant differences). However, a slight increase/decrease was observed in average nitrates/nitrites levels, respectively, which is attributed to the acidic pH of PAW, leading to the decomposition of nitrites into nitrates and nitrogen oxide.

Effect of PAW on Baby Spinach Leaves – Total Bacterial Counts

Right after the rinsing, total bacterial counts on PAW-treated samples were significantly lower than those achieved on untreated samples, although statistically significant differences were not found with regards to control samples rinsed with distilled water. It is noteworthy that although the indigenous bacterial flora on the control samples was reduced after rising with distilled water, total bacterial counts after 8 days of incubation at 4 °C were significantly higher than the bacterial load remaining on both PAW-treated and untreated samples. Despite the statistical analysis not yielding significant differences, the average bacterial concentration in PAW-treated samples was lower than the levels in untreated samples. Thus, about 1 log reduction in total bacterial counts has been achieved through PAW rinsing, with microbial levels resulting unaffected (5 log CFU/g) after 8 days of storage at 4 °C (typical shelf-life at retailers), which leaves a relative margin for shelf-life extension beyond current cut-offs at retailers, thus contributing to reduced industry and household food waste generation.

Effect of PAW on Baby Spinach Leaves – Surface Colour

Right after rinsing (day 0), no significant differences were found in the b* value (yellowness) among the three assayed conditions (untreated, PAW-treated and control with distilled water). However, significant variability was observed for the a* and L* values (redness and lightness, respectively), likely due to biological variability in the individual leaves analysed. It is noteworthy that average a* and L* values for PAW-treated samples were closer to the corresponding ones for untreated samples, as compared to the controls.

After 8 days of storage at 4 °C, no significant differences were found in the a* and b* values for PAW-treated and control samples. The L* value of the control samples was found to be similar to the untreated samples and significantly lower than that of the PAW-treated samples, but the average difference in the lightness of both treatments was limited to one unit, well within the range exhibited right after rinsing (day 0). On the other hand, the leaves subjected to either PAW or distilled water rinsing exhibited significantly lower and higher a* and b* values, respectively, than those for the untreated samples, due to relatively more pronounced yellowness and larger deviation from the initial greenness in the colour of rinsed samples as compared to untreated ones, as a result of storage. Such a difference is also attributed to the significant increase in greenness observed in the untreated samples at day 8, as compared to the colour obtained at day 0, characterized by significantly higher and lower a* and b* values, respectively.

These results demonstrated that rinsing steps influenced colour development of spinach baby leaves to a greater extent than the PAW treatment itself during chilled storage. In fact, the yellowness of PAW-treated samples remained unaffected during the 8 day chilled storage, while a significant increase was recorded for the control samples.