Possibility to extend the shelf life of NFC tomato juice using cold atmospheric pressure plasma

Not-from-concentrate (NFC) juices are premium, minimally processed products valued for high levels of bioactive compounds. Tomato juice, widely consumed for its taste and nutritional value (minerals, carotenoids, flavonoids, vitamin C), has a very short shelf life (typically one day) due to the absence of heat treatment and consequent microbiological instability, posing spoilage and safety concerns. There is a need for alternative, non-thermal processing to extend shelf life while preserving nutritional and sensory quality. Cold atmospheric plasma (CAP), an ionized gas rich in reactive oxygen and nitrogen species (RONS), offers antimicrobial efficacy without thermal damage and without chemical residues. This study investigates whether nitrogen-based CAP applied via a Glide-arc device can improve microbiological quality, extend refrigerated shelf life of NFC tomato juice, and preserve physicochemical properties and microstructure.

Prior studies report CAP’s antibacterial and antifungal activity in juices with minimal quality changes. HVACP reduced Salmonella in pasteurized orange juice with lower vitamin C loss than heat pasteurization (Xu et al.). DBD-CAP inactivated E. coli in apple juice by ~4 log with slight effects on °Brix, pH, acidity, color, phenolics, and antioxidant capacity (Liao et al.). CAP reduced bacterial counts in orange, apple, and tomato juices (Dasan et al.). CAP inactivated Zygosaccharomyces rouxii and Citrobacter freundii in apple juice (Xiang et al.; Surowsky et al.) and reduced Saccharomyces cerevisiae in white grape juice (Pankaj et al.). Plasma exposure may alter bioactive compounds depending on conditions; overexposure can degrade them. CAP-generated radicals can react with sugars, reducing fructose/glucose in orange juice (Almeida et al.). Mechanistically, CAP induces oxidative stress via RONS, damaging DNA, proteins, and lipids, leading to microbial death. Time-dependent plasma–liquid chemistry and pH modulate RONS species and antimicrobial efficacy.

• Raw material and juice preparation: Fresh tomatoes (Lycopersicon esculentum cv. Apis F1) from a biologically protected farm (Lublin Province, Poland) were washed, rinsed, dried, and pressed using a Sana EUJ-707 single-screw press (Omega Products, South Korea); skin and seeds were automatically removed. Juice was divided into six parts (control and CAP-treated) for microbiological, physicochemical, and microstructure analyses.
• Plasma treatment system: A Glide-arc (GAD) atmospheric-pressure plasma reactor with two 10-cm copper wire electrodes was used. Nitrogen gas (purity 6.0) at 440 L/h fed the reactor (50 Hz, 3.8 kV peak, 40 W). The exhaust gas was directed onto 50 mL of juice at 25 °C for 0 (control), 30, 60, 120, 300, or 600 s. Samples were in open glass containers (diameter 6 cm; wall height 4 cm) on a magnetic stirrer (50 rpm) in an ice bath. Ambient air mixing was possible; after 600 s, juice temperature rose to a maximum of 29 °C. Each of the six 50 mL samples was prepared in duplicate; for microbiology, n=5 per treatment time (three independent cultures from the first replicate and two determinations from the duplicate).
• Microbiological analysis: Post-treatment, samples were stored refrigerated in sterile containers for 1, 4, 7, and 10 days, then transported cold to an accredited lab. Serial dilutions were plated on appropriate media to enumerate total aerobic mesophilic microorganisms, lactic acid bacteria, coliform bacteria, and yeasts using standardized methods (PN/ISO). Pathogens Listeria monocytogenes and Salmonella were checked; E. coli contamination relative to regulatory thresholds was assessed. Counts are reported as log10 CFU/g.
• Physicochemical analysis: pH measured at 20±1 °C with a Metrohm 780 pH meter (calibrated pH 4–9). Total soluble solids (°Brix) measured at 20±1 °C with an Atago PAL-1 refractometer. Total carotenoids determined spectrophotometrically (Thermo Scientific UV-Vis Helios Omega 3) at 470 nm using a modified González-Casado method. Ascorbic acid determined by Tillmans titration with 2,6-dichlorophenolindophenol in oxalic acid.
• Microscopic analysis: Optical digital microscopy (KEYENCE VHX 950F) of fresh and CAP-treated samples (30–600 s) to assess parenchyma cell integrity and chromoplasts (lycopene crystals).
• Statistics: ANOVA with Tukey test at p<0.05 (Statistica v10). Microbiological results are means of five measurements; physicochemical properties are means of three measurements; whiskers denote SD in figures.

• Pathogens: No Listeria monocytogenes or Salmonella detected in any sample. E. coli was below the quantification limit (<10 CFU/g) throughout.
• Control (no CAP), refrigerated storage: Total aerobic mesophiles were 3.1 log10 CFU/g (day 1), 3.2 (day 4), 5.0 (day 7), 5.8 (day 10). Lactic acid bacteria: 1.9 (day 1), 2.8 (day 4), 3.7 (day 7), 5.6 (day 10). Coliforms: 0.6 (day 1), 1.2 (day 4), 2.7 (day 7), 2.8 (day 10). Yeasts: 0.6 (day 1), 2.6 (day 4), 2.6 (day 7), 3.6 (day 10).
• CAP 30 s: Minimal reductions vs control; decreases typically ≤0.2 log10 CFU/g across groups; no substantial shelf-life extension.
• CAP 60 s: Moderate reductions vs control after 10 days: total aerobes −0.3 to −2.4 log; lactic acid bacteria −0.3 to −2.4 log; coliforms up to −1.4 log; yeasts up to −1.2 log.
• CAP 120 s: After 10 days vs control: total aerobes −2.3 log; lactic acid bacteria −2.2 log; coliforms −2.2 log; yeasts −1.2 log. Slight microbiological improvement; shelf-life extension uncertain.
• CAP 300 s: Marked improvement; shelf life extended to 10 days per pasteurized juice criteria (aerobes and yeasts within allowable limits). Vs control: total aerobes −2.4 log (day 7) and −3.3 log (day 10); lactic acid bacteria −1.4 log (day 7) and −5.0 log (day 10). Coliforms and yeasts reduced below quantification (<10 CFU/g; <1 log10) at all storage times.
• CAP 600 s: Greatest effect; shelf life extended to 10 days. After 10 days: total microorganisms at 1.6 log10 CFU/g; lactic acid bacteria, coliforms, and yeasts all <10 CFU/g (<1 log10). Reductions vs control after 10 days: total aerobes −4.2 log; lactic acid bacteria −5.0 log.
• Time-dependent inactivation: Greater apparent reductions after storage reflect control growth vs declining counts in treated samples; literature reports similar delayed inactivation due to longer-lived RONS (e.g., H2O2, OOH−, ONOO−).
• Physicochemical properties:
  • pH: Control ~3.99 (day 1). CAP caused slight increases over storage. For 30–300 s, pH ~4.44–4.51 over 1–10 days. For 600 s, pH 4.47 (days 1–4) rising to 4.94 (day 7) and 4.90 (day 10).
  • °Brix: Control 3.65 °Brix (day 1). CAP-treated samples 3.75–4.25 °Brix, with slight increases; 600 s reached up to 4.25 °Brix (day 7). Storage time had no significant effect overall.
  • Carotenoids: Control ~77.5 mg/100 g (day 1). CAP slightly increased levels early; maximum 79.15 mg/100 g (600 s, days 1–4). Over storage, slight declines, but treated samples generally remained similar or marginally higher than control; lowest observed 74.25 mg/100 g (30 s, day 10).
  • Ascorbic acid (vitamin C): Control 12.27 mg/100 g (day 1) decreasing to 11.70 by day 4 and 9.09 by day 7 (table shows up to day 7 for control). CAP treatments largely maintained vitamin C; 600 s showed only slight losses (max ~5%), with values around 11.69–11.75 mg/100 g over 1–10 days. Storage had minimal impact in treated samples.
• Microstructure: Digital microscopy showed mostly intact parenchyma cells and chromoplasts (lycopene crystals) after CAP, including at 300–600 s; only slight chromoplast damage (amorphous red pigment clusters) observed at longer treatments.

The study demonstrates that nitrogen-based CAP generated by a Glide-arc reactor can substantially reduce natural microflora in NFC tomato juice and extend refrigerated shelf life without significant alterations in key quality metrics. The antimicrobial efficacy increased with treatment time, with 300–600 s achieving reductions sufficient to keep total aerobes and yeasts within or below thresholds aligned with pasteurized juice criteria over 10 days of storage. Differential susceptibility was observed: coliform bacteria were highly sensitive (counts <10 CFU/g at ≥120 s), lactic acid bacteria were markedly reduced at longer treatments (notably 300–600 s), whereas yeasts required the longest exposures for sustained suppression. Observed greater reductions after storage are consistent with time-dependent plasma–liquid chemistry: short-lived species generated during discharge give rise to longer-lived RONS (e.g., H2O2, peroxynitrite derivatives) that persist and continue inactivation. Juice acidity likely enhances antimicrobial action by modulating species such as HO2 and O2NOH that more readily permeate cells at lower pH. Importantly, physicochemical properties (pH, °Brix, total carotenoids, vitamin C) were largely preserved, with only slight pH and °Brix increases and minimal vitamin C loss (≤~5% at 600 s). Microscopy confirmed structural integrity of juice cells and chromoplasts with only minor damage at longer treatments. Overall, CAP effectively addresses the central challenge of extending NFC tomato juice shelf life while maintaining nutritional and structural quality, and suggests feasibility for industrial-scale, chemical-free decontamination.

Nitrogen-based cold atmospheric plasma applied via a Glide-arc device for 300–600 s effectively decontaminated NFC tomato juice, extending refrigerated shelf life to 10 days while largely preserving physicochemical properties and microstructure. A 300-s treatment significantly improved microbiological quality; 600 s achieved further reductions, bringing lactic acid bacteria, coliforms, and yeasts below the quantification limit. Microscopy indicated mostly intact cellular structures post-treatment. These findings support CAP as a promising, waste-free, non-thermal technology for industrial processing of NFC tomato juice. Future work should elucidate biochemical and molecular mechanisms of microbial recovery or death post-CAP and optimize process parameters for different juice matrices.

The study was performed on one tomato cultivar and a single CAP configuration (nitrogen-fed Glide-arc) and evaluated a limited set of microbial groups by culture-based methods. Mechanisms underlying observed microbial recovery or delayed inactivation were not investigated; the authors note that biochemical and molecular studies are needed to explain these effects. Additionally, while physicochemical and structural parameters were measured, sensory evaluation was not reported, and RONS concentrations in the liquid phase were not quantified.