Skip to main content

Novel continuous in-pack decontamination system for fresh produce

Final Report Summary - SAFE-BAG (Novel continuous in-pack decontamination system for fresh produce)

Executive Summary:
SAFEBAG was a three year applied research project, which commenced in September 2011 and finished in October 2014, and was funded under the "Capacities/Research for SME-AGs" programme of the European Commission´s Seventh Framework Programme (FP7).
The EU-27 fruit and vegetable processing and supply chains represent a major pillar of the European food and drink industry. While fresh fruit and vegetable consumption is linked to a plethora of health benefits, it can also be a source of foodborne illnesses and as such we are witnessing an increase in the number of outbreaks associated with ready-to-eat fruit and vegetables. The treatments currently used, including chlorine washing, often leave a chemical residue and waste water. There is growing demand to reduce the amount of chemicals used in the process. This can be achieved through an effective yet environmentally-friendly decontamination system based on plasma.
Plasma is electrically energised gas whereby constituent molecules of the gas split to yield free electrons, ions, quanta of electromagnetic radiation, etc. There are several types of plasmas depending on the conditions in which they are generated. Pre-project research conducted by Dublin Institute of Technology indicated that in-pack atmospheric cold plasma (ACP) could significantly reduce the microbial load of fresh fruits and vegetables. The SAFEBAG project has advanced this knowledge by carrying out further research to develop a novel continuous in-pack decontamination system for fresh-cut produce and to maximise its impacts.
In the SAFEBAG approach, the food package is treated between two high voltage electrodes. The high-voltage process ionises the gas within the electric field, including the gas contained within the package. It is this mix of active species which results in the anti-microbial effect. Crucially the active gas reverts back to stability shortly after, meaning no residual chemicals are left on the product. An important novelty of this technology lies in generating the plasma inside a sealed package containing the produce, which facilitates rapid treatment and eliminates the risk of post-process contamination.
A pre-competitive prototype has been designed and built, and its effectiveness for decontaminating fresh bagged fruits and vegetables validated under industrial conditions. SAFEBAG is a dry, non-thermal and chemical-free washing technology, compatible with online production and MAP packaging, which leaves no hazardous residues in the treated produce. The developers of technology have been committed to making this system affordable, robust and easy to maintain.
The project has proven the SAFEBAG concept and demonstrated the system to the industry, however the results are at precompetitive scale, and further scale up work and demonstration effort is required now in a post-project phase. Such developments could make the technology accessible to European fresh-cut SMEs, whereby they will benefit from being able to differentiate their products and market them as a value added commodity. The impact of the results hold benefits for consumer safety and confidence, extended shelf-life and increased demand for fresh produce, which will in turn impact on the competitiveness of hundreds of European fresh-cut processing SMEs.

Project Context and Objectives:
The overall objective of the project was to develop a prototype of a novel in-pack plasma-based process to reduce microbial load in fresh produce, thereby ensuring food safety and extending shelf-life without compromising nutritional or quality parameters.
In order to achieve the above, the technical and operational objectives that needed to be fulfilled are provided below.
1. To build a lab-scale NTP system that will enable each of the RTDs laboratories to investigate the conditions of operation and to help facilitate scale-up.
2. To evaluate the impact of the requirements of the European Food Safety Authority regulations on novel technologies on the proposed SAFE-BAG system.
3. To implement a ‘bottom-up’ approach whereby the needs and specifications of European lettuce, fruit and vegetable packers will be consulted and the findings used to define the specifications for the SAFE-BAG system.
4. To carry out quality tests on the foods that have been treated with the NTP test-rig at laboratory scale, in order to better optimise the process and to ensure that the technology can be validated for compliance with novel technology regulations.
5. To characterise the physical plasma discharge produced under the variable control parameters employed using specific plasma diagnostics techniques and to correlate plasma discharge characteristics with anti-microbial efficacy
6. To design and build a precompetitive SAFEBAG prototype for the in-pack decontamination of fresh lettuce, fruit and vegetables taking into account the inputs from WP2 and WP3. (WP4, achieved).
7. To install, test and validate the SAFEBAG system in industrial facilities, to run in-pack decontamination trials with the system and to perform safety, shelf-life and quality analysis of the treated samples (WP5 and WP6, partially achieved).
8. To outline scaling-up guidelines and development work for full production (WP5, achieved).
9. To carry out a comprehensive series of demonstration activities proving the viability of the prototype, and outlining its potential productivity, quality enhancing and economic advantages.
The overriding goal of this project was to ensure that the pre-competitive SAFEBAG prototype resulting from this project fulfils the threshold requirements to ensure its further development post-project into a fully industrial system that is taken to market.

Project Results:
The project started WORK PACKAGE 1 by understanding of the technological needs of European fresh-cut F&V packers, as well as market needs and perceptions in terms of product quality and safety.
A survey was carried out among fresh and fresh-cut F&V manufacturers/suppliers across Europe, in order to gain an understanding of their activities, technological needs etc. In total 20 in-depth consultations (6 field visits and 14 phone interviews) were made to fresh and fresh-cut F&V industries in Spain, Ireland, Slovenia, Austria, Czech Republic, United Kingdom and Turkey.
A comprehensive literature and patent review was in order to update the state-of-the-art in terms of disinfection and plasma technologies and to identify relevant patents that must be considered in order to define the specifications of the system.
The safety of European consumers is paramount and therefore the Novel Foods Regulation (EC 258/97), which legislates the application of new and emerging novel technologies, was also studied in detail. The procedures involved in ensuring that the SAFE-BAG technology complies with EC regulations were analysed.
The overall system specifications and performance characteristics of the SAFE-BAG system in meeting the needs and requirements of the fresh-cut F&V sector were defined, which provided a very solid platform of knowledge for the project.

WORK PACKAGE 2 focused on evaluating critical control parameters relating to antimicrobial efficacy of ACP, with tandem evaluation of those critical parameters for quality testing of ACP-treated foods, towards the combined goals of promoting microbiological safety and quality characteristic retention over an extended shelf-life for fresh produce. T
Two modular based test rigs were designed and built, including voltage transducers of 60 and 120 kV. The test rigs designed were tested for antimicrobial efficacy against range of microorganisms.
In order to determine the microbial inactivation capacity of the designed ACP systems, critical control parameters of voltage level, type of gas, mode of exposure, treatment time, relative humidity as well as post-treatment storage time were mainly investigated. The effect of these parameters on inactivation rate was investigated using the two ACP systems i.e. DIT 60 and DIT 120.
1. Post-treatment storage time emerged as a critical treatment parameter for consistency of bacterial inactivation. A reduced treatment time (60 s) in conjunction with 24 h of post treatment storage time showed an enhanced plasma inactivation effect with complete bacterial inactivation.
2. The mode of plasma exposure has effects on antimicrobial inactivation capacity of the ACP systems, however, there are interactive effects with other parameters, and the decision to use direct or indirect plasma exposure may be related to products sensitivity for quality characteristic retention.
3. This study indicated that the microbial inactivation capacity of the two systems DIT60 and DIT120 i.e. application of 40 kV peak to peak or 56kVRMS respectively, appeared to be different. Although, both ACP systems evaluated showed significant bacterial reductions, a strong impact of increasing voltage level on inactivation of microorganisms was noted.
4. It can be concluded that utilisation of a gas mixture with higher oxygen content for ACP generation results in efficient bactericidal effects with greater production of ozone.
5. It was observed that optimised parameters altered the cell surface structure, with release of cell components, thus affecting cell integrity, with DNA damage depending on type of bacteria tested.

The efficacy of ACP for decontamination treatment of a range of selected produce types was evaluated. The previously optimised control parameters of voltage level, inducer gas and mode of exposure were mainly examined for inactivation of E.coli and Salmonella spp microorganisms on fresh produce surfaces including cherry tomatoes and strawberries. This study clearly demonstrated that increasing the treatment time always resulted in increased decontamination efficacy of ACP treatment. However, influence of produce surface structure and bacteria type on ACP antimicrobial efficacy was observed. Salmonella appeared to be more sensitive to ACP treatment than E. coli when inoculated on smooth surface of tomatoes.
Mechanism of action studies were carried out using DIT 120 system. The effect of optimised ACP parameters on bacterial cell surface structure, cell integrity and DNA damage was mainly studied employing methods such as scanning electron microscopy, absorbance measurement and Polymerase chain reaction (PCR). It was observed that optimised parameters altered the cell surface structure, with release of cell components, thus affecting cell integrity, with DNA damage depending on type of bacteria tested. Using higher voltages, L. monocytogenes appeared to be more sensitive than the two strains of E. coli studied. It also showed more damage to genomic DNA and less PCR product formation on agarose gel.
Quality changes of ACP-treated produce were examined. In order to determine effects on quality parameters three produce types were selected i.e. cherry tomatoes, strawberries and spinach.
1. In-package ionisation of cherry tomatoes as a surface treatment process did not significantly affect the instrumental and objective quality parameters.
2. In-package ACP processing did not induce significant physiological (respiratory) stress or adversely affect the colour and firmness of strawberries.
3. ACP treatment of baby spinach leaves caused adverse change in texture and wilting, irrespective of whether direct or indirect treatments were carried out.

The application of ACP requires consideration of the packaging material utilised and the potential for interactive effects on microbiological safety retention and toxicological safety profiles of the packaging material that may occur. Thus, the changes in the polymer film surface and bulk property, along with its compliance to the legal regulations for food safety were assessed.
1. upon ACP treatment was observed in relation with the treatment time.
2. ACP treatment also lead to an increase in the WVTR of the film.
3. While the overall migration increased after ACP treatment of different food simulants, the migration remained less than the limit provided by EU regulations for food packaging materials.
Based on the results obtained, it was decided that the DIT 120 kV system would be employed as the basis of the precompetitive prototype to be built by IRIS in WP4, given its high voltage output, control features and duty cycle, facilitating large gaps to be treated under continuous operation. Design and control recommendations were made for scale up including, voltage, electrode design and gap, frequency and dielectric materials.

The objective of WORK PACKAGE 3 was to characterize the physical discharge and the induced plasma chemistry produced by ACP, aiming to identify the reactive species key to bacterial inactivation.
The ACP produced under the variable control parameters (applied voltage, gap width, dielectric thickness, gas type, gas humidity levels) was characterised using electrical methods (charge-voltage measurements) and emission or photo-absorption spectroscopy, respectively.
The obtained data and correlation of the plasma discharge characteristics with anti-microbial efficacy showed that:
1. The main plasma reactive species involved in bacterial inactivation can be identified as ozone and nitrogen oxides at high and low densities , respectively;
2. The ozone concentration dependence for several applied voltages, gaps, dielectric thickness, and gas compositions including humidity levels and the correlation to antimicrobial effects;
3. The concentration of ozone in the closed package reaches uniform distribution during the plasma exposure times required for sterilization (15 - 120 s) allowing recording of maps of ozone concentration versus plasma-on time.
4. Better accuracy levels in absorption spectroscopy data will allow for determination of nitrogen oxides levels and possibly to a correlation of their concentration levels against antibacterial effects.

A Comsol model of the DBD discharge in Ar, nitrogen and oxygen pure gas was set-up, including the gas-phase chemistry and boundary conditions. Regarding boundary conditions involving the secondary electron emission generated at the electrodes (dielectric surfaces) through sputtering or metastable Auger de-excitation on surfaces - a comprehensive study was performed for each gas considering a range of possible energy and yields for electrons generated by metastables. The results indicated that the N2 and O2 discharges are quickly depleted of electrons once gas dissociation takes place and a significant contribution from secondary electron emission at electrodes is required to maintain the discharge. This is in agreement with known DBD operation in N2 and O2 gas and indicates the development of a filamentary discharge where the ratio between the first (α) and second (γ) Townsend coefficients is below 1 (α/γ <1). Only the DBD plasma in Ar can maintain a typical Townsend glow discharge where α/γ >1.
Code for advanced spectral de-convolution based on the data on optical absorption measurements was developed. The advanced deconvolution showed the following major components of humid air after plasma exposure (O3, NO2, NO3, N2O4 and N2O5) function of the RH levels and exposure time. Other components are inferred from the de-convolution residual and appear to be HNO4 and NO2 on exited states.
The convergence of the code for synthetic air and by setting-up was improved and several reaction-sets for DBD discharge and plasma chemistry simulations in humid air evaluated. The time convergence for synthetic air was improved up to 40 ms through improving the set of γ coefficients.
Once built, the SAFEBAG prototype was tested recording the optical emission (OES), discharge current and amounts of ozone, dissociated water and hydrogen present in the plasma-exposed bags as a function of operation parameters and post-discharge time. An exponential decrease in measured species concentrations was observed with values of ozone and hydrogen still high (> 500 ppmv) at 30 minutes in the post discharge.
Several chemical reaction sets for humid air plasma chemistry were tested (including 77 - 160 reactions) and a compelling set of COMSOL numerical experiments performed to allow for the code convergence. This was difficult to achieve due to the extreme complexity of the time-space dependent problem and wide time-scales involved (10-9 seconds up to several minutes). A segregation of the discharge development and plasma chemistry has not been possible within the framework of COMSOL-DBD model and a different approach has been taken.
A Global Model of the discharge based on using the BOLSIG code simulations in conjunction with a time-dependent chemistry solver routine was developed. The BOLSIG code uses as input the measured transferred power and current in the discharge generating the necessary plasma parameters for the calculation of reaction rates in the humid air plasma; which can be used for initializing the plasma chemistry at each time-step in the reaction set chemistry routine. This approach ensures the convergence of the code for times up to 2 minutes during discharge.
The simulations for the post-discharge (up to 2 minutes) to allow for a comparison to experimental data on plasma chemistry from absorption spectroscopy were continued. The results confirm the high ozone levels obtained at 70kV applied voltage and the decrease in O3 levels with increase in air RH with formation of HO2, H2O2 (peroxides) and HNOx (x=1, 4) especially, the H2O2 and HNO3 and HNO4 persisting at high concentrations in the post-discharge (100’s of ppm). Such results clarify the intense bactericidal effects of plasmas in humid air and closed containers showing the synergetic effects of ozone, peroxides, nitrogen oxides and their acids.

The main objective of WORK PACKAGE 4 was to design and build the pre-competitive prototype for the in-pack decontamination of fresh-cuts, in keeping with the results obtained at laboratory level (WP2 and 3) and taking into account the requirements described by end-users of the project and a wider community of European F&V processors (WP1).
In T4.1 the first step was to draw a simple block diagram describing the main functions of the prototype as agreed in D1.1.Based on the results obtained in WP2 and WP3, DIT and DCU made scale-up recommendations including voltage, electrode design and gap, frequency and dielectric materials. It was decided that the DIT 120 kV system would be employed as the basis of the precompetitive prototype, given its high voltage output, control features and duty cycle, facilitating large gaps to be treated under continuous operation. Based on these recommendations, several design options were proposed to provide the necessary functionality, which resulted in a function means analysis. From this, the simple block diagram was completed with detailed parts’ specifications in order to generate a final block diagram.
In short, the SAFEBAG system can be described as follows: two side-grip belts displace sealed bags filled with fresh-cut produce into the treatment zone, where they keep moving along the gap between two electrodes that are connected to a high voltage transformer. When contact between the food package and the electrode occurs, a dielectric barrier plasma discharge takes place in the gap between the two electrodes. This high voltage process ionises the gas contained within the food package, resulting in the generation of significant amounts of reactive species that have a bactericidal effect on the fresh-cut product. The speed of the belts determines the duration of the treatment, after which the bags are released onto the unloading platform.
The SAFEBAG prototype was be built using the detailed designs earlier. Some components (e.g. high voltage power supply) were purchased off-the-shelf, whereas others were manufactured at IRIS. The different modules that make up the system were integrated into a unique system. The control and supervision of the entire system were designed in order to render the machine as automatic as possible: Different control strategies were studied and the control loops designed. Sensors, actuators, timers, controllers, probes and the rest of items required were selected and purchased. In parallel, several technologies for the supervision part of the machine were studied and implemented in the prototype: electrical, optical and acoustic. Such supervision allows for measurements in real time, giving the needed information to the user (display in the control panel) or a more exhaustive and technical information to the technician (via a connection to a computer or/and an oscilloscope directly to the BNC connector in the control panel).
The objective of WORK PACKAGES 5&6 was to install and validate the SAFEBAG prototype in conditions that are representative of industrial practice. The trials took place at DIT´s pilot plant (Dublin, Ireland) with examination of system efficacy in terms of food safety profiles, shelf-life extension and effects on quality and nutritional parameters. The prototype SAFEBAG system was operated under both static and motion conditions to represent fresh produce packages post sealing in the production line. The reason for including static conditions in this study was to reflect batch conditions, which had been used for testing the test-rig in WP2. The produce considered included strawberries, cherry tomatoes and spinach.
The antimicrobial efficacy of the SAFEBAG prototype for inactivation of background and pathogenic microorganisms on fresh produce was investigated with subsequent storage under refrigerated conditions. Useful reductions in pathogenic and background microflora were achieved with slightly enhanced efficacy using static mode of treatment rather than continuous for all samples. Significantly better results were observed for strawberry and cherry tomato products than for spinach which could be linked to product characteristics including differences in surface structure and the ability of microorganisms to adhere or internalise.
Parallel studies examined any potential changes in the quality and nutritional attributes of fresh produce after being treated SAFEBAG prototype system. These parameters included colour parameters, texture, pH, TSS and firmness. Several instrumental techniques were used to characterise the effects of the treatment on the quality and nutritional profiles of plasma treated packages. In-package ACP treatment of strawberries, cherry tomatoes and spinach was found to not adversely affect the critical quality parameters of colour, pH, total soluble solids and firmness.
The potential of the prototype to extend the shelf like of product was also examined. The antimicrobial efficacy on the background microbial populations of samples produce was studied over the products shelf life. Background microflora of the produce was measured, as the level of natural microbial load is one of the major factors causing fresh products’ deterioration. The surviving background microflora including estimations of aerobic mesophilic bacteria along with estimates of yeasts and moulds were evaluated. The ability of the prototype to extend the shelf-life of fresh produce was investigated with respect to microbial growth of spoilage organisms over a storage period of up to 9 days.
The SAFEBAG project has taken the development to pre-competitive level. The results of the tests carried out indicate that the SAFEBAG system can be used as a tool to treat fresh-cut produce, however technical challenges have also been met over the validations period. Further post project development and demonstration work is needed to scale up to a commercial system that is ready for market launch. Detailed information regarding additional development work required to industrialise the project outputs has been documented.

The main objective of WORK PACKAGE 7 was to carry out a comprehensive series of demonstration activities in order to prove the industrial viability of SAFEBAG, to highlight its technical performance features, as well as to outline its potential and limitations.
The SAFEBAG project consortium planned four demonstration events of project results and achievements to public. The demonstration sessions were prepared as workshops that included an introduction to the project, a summary of the results obtained in the laboratory and the results obtained with the SAFEBAG prototype in terms of microbial load reduction, nutritional and quality profile retention and shelf-life extension of fresh-cut F&Vs. Practical demonstrations of the operational prototype were also planned.
Potential Impact:
The SAFEBAG project has developed a pre-competitive prototype of a novel continuous in-pack decontamination system for fresh-cut produce, based on atmospheric cold plasma technology. SAFEBAG is a dry, non-thermal and chemical-free washing technology, compatible with online production and MAP packaging, which leaves no hazardous residues in the treated produce. The developers of technology have been committed to making this system affordable, robust and easy to maintain.

The end-users of the technology will be able to differentiate their products and gain competitive advantage, through:
• Increased safety profiles of fresh produce: In recent years there has been growing concern over the presence of foodborne pathogens in bagged salads. Access to an in-pack technology that will ensure a substantial reduction in microorganisms will add significant value to their product offering.
• The enhancement of microbiological and organoleptic quality parameters by comparison with current processing protocols will lead to shelf-life extension: The major obstacle of purchasing ready-to-eat fresh-cut fruits and vegetables is their short shelf-life, leading to quick degeneration and decomposition of the product and undesirable look and negative palatability.
• Reduced water usage: fresh produce washing is a major consumer of potable water, and water-pricing policies with incentives for efficient water use are being implemented under EU´s Water Framework Directive. A dry preservation technology will allow a considerable reduction in water as well as wastewater generation.
• Replacement of chlorine: there is a need to eliminate/reduce chlorine from the disinfection process because of its effectiveness and the concerns for the environment, as well as health risks.
By having access to a technology such as SAFEBAG, fresh-cut fruit and vegetable suppliers will be equipped to provide products that deliver on safety, taste and freshness. This will result in an increased confidence in ready-to fresh-produce by the consumers, which will in turn impact on the competitiveness of hundreds of European fresh-cut processing SMEs.
Key importance was given to the management of the intellectual properties and in agreement of the dissemination of non-confidential information throughout the project. A preliminary business plan, as well as a post project development word, have been laid out. The synergistic role of the partners covering the whole supply and value chain has been discussed. Successful dissemination activities were carried out on the principles of the SAFEBAG technology, through the project website, leaflet and poster; a number of scientific articles and attendance to conferences, meetings and trade fairs both in industry and in the public domain.

List of Websites:
www.safebag.eu

Project Coordinator:
Dr. Edurne Gaston Estanga (egaston@iris.cat) on behalf of
Innovació i Recerca Industrial i Sostenible
Avda Carl Friedrich Gauss nº11
08860 Castelldefels
Spain
+34935542500