Pulsed Electric Field Processing of Orange Juice: A Review on Microbial, Enzymatic, Nutritional, and Sensory Quality and Stability
Direct inquiries to author Buckow (E-mail: firstname.lastname@example.org).
During the last decades pulsed electric field (PEF) processing received considerable attention due to its potential to enhance food products or create alternatives to conventional methods in food processing. It is generally acknowledged that PEF processing can deliver safe and chill-stable fruit juices with fresh-like sensory and nutritional properties. Relatively low-processing temperature and short residence times can achieve highly effective inactivation of microorganisms while retaining product quality. A first commercial application of PEF for preservation of fruit juices was launched in 2006 in the United States. Since then, industrial-scale processing equipment for liquid and solid products were developed and, in Europe in 2009, an industrial juice preservation line was installed using 20 kV/cm pulses at 40 to 50 °C to extend the chill-stability of fruit juices, including citrus juices and smoothies, from 6 to 21 d. The related PEF processing costs are in the range of US $0.02 to 0.03 per liter and are justified due to access to new markets and reduced return of spoiled product. However, despite its commercial success there are still many unknown factors associated with PEF processing of fruit and citrus juices and many conflicting reports in the literature. This literature review, therefore, aims to provide a comprehensive overview of the current scientific knowledge of PEF effects on microbial, enzymatic, nutritional, and sensory quality and stability of orange juices.
Pulsed electric field (PEF) processing has the ability to effectively inactivate microbial cells, when combined with low to moderate temperatures (<50 °C), which makes it a promising alternative to conventional thermal preservation processes for liquid foods that contain heat labile bioactive or volatile components such as fruit and vegetable juices. The sensitivity of microorganisms to PEF treatments depends on cell characteristics such as structure and size (Toepfl and others 2006). In addition, factors such as product pH, water activity (aw), soluble solids, and electrical conductivity affect the efficiency of the technology to affect biochemical reactions and inactivate microorganisms (Aronsson and Rönner 2001).
Although the underlying mechanisms are not yet fully explained on a molecular basis, PEF treatment results in changes in the membrane permeability of biological cells (Gásková and others 1996). This effect can be exploited to inactivate microorganisms or to enhance mass transfer in extraction or drying processes.
Typically, food is placed between 2 electrodes and exposed to an electrical field in the form of very short (a few μs), high-voltage (kV) pulses. One electrode is connected to a high-voltage switch and the other to the ground. The electric field strength E generated between the pair of electrodes can be estimated by dividing the applied voltage U by the distance between the electrodes d (that is, E = U/d).
The electric field strength, treatment temperature, treatment time, and specific energy input are the main processing parameters affecting the degree of microbial inactivation (Alvarez and others 2006). To ensure effective microbial inactivation, electric field strengths should be in the range 20 to 50 kV/cm, pulse lengths in the range of 1 to 10 μs and specific energy inputs in the range 50 to 1000 kJ/kg (Toepfl and others 2007). However, the multitude of variable parameters of PEF technology and the food characteristics often requires a systematic study of the individual influence of these parameters on desired and undesired reactions.
To establish a typical PEF system, a pulse generator is needed. A typical system comprises a high-voltage power supply, one or more (energy storage) capacitors, a high-voltage switch and a treatment chamber. Furthermore, a liquid handling system including heat exchangers, a product pump, and monitoring devices (oscilloscope, temperature sensors) to control and report the conditions, are necessary (Zhang and others 1995). To deliver the energy at high voltages within a short period of time (μs) a capacitor is needed. The capacitor has the function of storing the generated energy from the pulse generator at the selected voltage. When the high-voltage switch is closed, energy is discharged from the capacitor almost instantly to the electrodes in the form of a high-voltage pulse of a few microseconds. The relatively slow charging of the system compared to the fast high-energy discharge of the capacitors often requires significantly longer time spans (about 100 to 1000 times) between the pulses than the width of the pulse itself.
The chamber design exhibits a significant influence on the effectiveness of the process by affecting treatment uniformity, peak electrical field strength, and product throughput (Buckow and others 2011). One challenge is to design a treatment chamber capable of operating at high and uniform electric field intensities and which prevents dielectrical breakdowns (Qiu and others 1998). Batch chambers with parallel plate electrode configurations provide relatively low throughputs but high-treatment uniformity. Treatment chambers with colinear configurations of electrodes allow continuous operation at high throughputs, but often exhibit poor treatment uniformity.
Dielectrical breakdowns occur when the applied electrical field strength exceeds the dielectric strength of the treated food product in the chamber (Zhang and others 1995). Dielectrical breakdowns can also be caused by local field enhancement and impurities (gas bubbles or solids) in liquid foods. Not less important are other relevant design criteria such as electrical resistance of the chamber, low-erosive electrode material (for example, platinum), improvement of flow behavior and temperature distribution, and of course the treatment capacity (that is, product throughput and constant power supply) (Jaeger and others 2009). Nonuniform electric field and flow velocity distributions can result in under- (often in central regions or dead spaces) or overtreated (often in boundary regions) volume elements, which lead to an increased chance of electrical sparks and system breakdown, as well as degradation of the quality of the treated product. Thus, the effectiveness and efficiency of the PEF treatment decreases.
The majority of pilot and industrial-scale PEF systems comprise treatment chambers (flow cells) with cofield and colinear configurations of electrodes. Such chambers have 2 or 3 hollow cylindrical electrodes (2 high-voltage electrodes and 1 grounded electrode) separated by hollow insulating spacers. The product flows through the formed tube. This electrode design and configuration provides a very nonuniform electrical field and an undefined treatment zone (Van den Bosch 2007). The nonuniformity may result in insufficient treatment, dead spaces and, thus, possibly recontamination of the treated medium with microorganisms. This presents a major challenge when the technology is used for pasteurization applications where a 5- or 6-log10 reduction of pathogens has to be achieved.
To reduce the risk of under- or overprocessing, multiple PEF treatment chambers can be placed in series. For example, in a PEF system that operates continuously (for example, using colinear treatment chambers), treatment uniformity can be enhanced by adding the number of treatment flow cells. This numbering up of treatment zones also means that the required processing time can be broken up into smaller fractions allowing intermediate cooling of the product. This can slightly reduce the effectiveness of the treatment but will preserve the quality of the product better than in the case where the required process intensity (that is, specific energy) is applied in one treatment zone, which often results in a significant temperature rise.
In general, a turbulent flow is desirable because it results in a more uniform residence time (Heinz and others 2002). The high intrinsic resistance of colinear chambers makes it possible to use several chambers in series and, in addition, enables the pulse generator to operate with smaller currents than in coaxial treatment chamber designs. Additional benefits of colinear systems include the possibility of a greater diameter enabling higher product throughput, the ease of cleaning, an effective usage of electrical energy, and good flow dynamics (Buckow and others 2010).
Another way of enhancing treatment uniformity of continuous PEF systems is to modify the dimensions and geometry of the treatment chamber, which can either result in a more uniform distribution of the electrical field strength and/or enhanced flow patterns of the treated food. Modification of PEF treatment chamber design can either include shape modifications of the insulator (Buckow and others 2011) or electrode configuration (for example, a concentric rift) (Toepfl 2011) or insertion of insulator material or a metal grid into the flow shapes configurations (Alkhafaji and Farid 2007; Jaeger and others 2009). However, current PEF research and development is mainly focused on further optimizing colinear treatment chamber designs rather than looking into new ways of applying high-voltage pulses to food systems (Huang and Wang 2009).
PEF Inactivation of Microorganisms in Orange Juice
Mechanisms of microbial inactivation by PEF
The inactivation of vegetative bacteria and yeasts during PEF processing is probably not due to the products of electrolysis or temperature rise alone, but rather determined by the applied electrical field strength and the treatment time (Evrendilek and Zhang 2003; Heinz and others 2003). There are several hypotheses on the mechanisms involved in the rupture of the cell membrane when exposed to an electric field. Two hypotheses, electrical breakdown, and osmotic imbalances, are widely accepted and are based on the same principles.
The theory of electrical breakdown considers the cell membrane as a capacitor filled with a dielectric medium (Zimmermann 1986). The cell cytoplasm has a greater dielectric constant (6 to 8 times) than the membrane, as does the liquid food that surrounds the cell. The difference between dielectric constants on either side of the membrane results in a transmembrane potential (TMP) of about 10 mV (Jeyamkondan and others 1999). The TMP is created through free charge accumulations at the inner and outer surface of the cell membrane. When an external electrical field is applied, ions inside and outside of the cell move along the field until they are restrained and accumulate at the membrane and an increase in the TMP occurs. The ions of opposite charge (+ and –) on either side of the membrane are attracted to each other, compress the membrane and reduce its thickness. When the electrical field strength is increased to a point where it exceeds a critical threshold value of the TMP, an electrical breakdown or pore formation is induced. The critical TMP value is around 1 V (Hamilton and Sale 1967). The electrical field strength required to achieve an electrical breakdown of the cell membrane depends on cell size and shape, cell orientation in the electric field, dielectric characteristics of liquid food, cytoplasm and membrane, and temperature.
Breakdown and pore formation can be reversible or irreversible, depending on the treatment intensity. If the intensity of the PEF is such that the energy deposited on the membrane does not result in considerable Joule heating the cell can often recover from the damage done to its membrane (Kolb and others 2006). In this case, the cell membrane can remain permeable for some minutes depending on the temperature after exposure to mild PEF treatment (Rols and Teissie 1998). However, when pores in the membrane surface become numerous and/or large in size, an irreversible breakdown of the membrane occurs, which leads to a mechanical destruction of the membrane and subsequent cell death.
The theory of osmotic imbalance describes the loss of cell membrane functionality through formation of hydrophilic pores in the membrane and the forced opening of protein channels. The electrical field causes changes in the conformation of phospholipids, leading to rearrangement of the membrane and formation of hydrophilic pores. The opening and closing of protein channels embedded in the membrane also depends on the TMP. There is evidence that charged H+ protons rapidly move along a lipid-water interface (for example, a biological cell in a suspension) and can affect transmembrane proton conduction (Teissie and others 1985). The gating potentials of protein channels that must be overcome for solutes (for example, ions) to flow through are approximately 50 mV, which is 150 to 500 mV less than the breakdown potential of the lipid bilayer of the membrane (Tsong 1991). During application of an electric field, protein channels might not only open, but also become denatured (irreversible pores) by local Joule heating or electric modification of functional groups (Tsong 1991).
There are reports that the conductivity of the media surrounding the biological cell will affect the permeabilization process during PEF treatment (Ivorra and others 2010; Moisescu and others 2013). This is because the time of charge accumulation on the cell surface is smaller in high-conductive media. This means that microbial inactivation is likely to be enhanced in high-conductive media than in a low-conductive environment (for example, <1 mS/cm). However, increasing the electrical conductivity also increases the flow of electrical current leading to pronounced Joule heating and increased energy consumption during PEF processing.
Factors affecting microbial inactivation
Many process parameters affect the effectiveness of the PEF treatment for inactivating microorganisms. Changing one parameter might affect another. These circumstances make it difficult to compare the results of different studies. Nevertheless, the main process parameters influencing the effectiveness of PEF treatments and, thus, microbial inactivation are the applied electric field strength, pulse length and shape, total treatment time, treatment temperature, and specific energy input (Heinz and others 2002).
Typically, the greater the electric field strength, higher the temperature or longer the treatment time, the greater the microbial inactivation (Wouters and others 2001). For example, a study by McDonald and others (2000) demonstrated the effects of PEF on microorganisms in orange juice applying electric field strengths of 30 and 50 kV/cm for 12 μs and at 55 °C outlet temperature. PEF treatment at 30 or 50 kV/cm inactivated 4.75-log10 and 6.2-log10 CFU/mL Leuconostoc mesenteroides in orange juice, respectively. Escherichia coli O157:H7 inoculated in orange juice was inactivated 1.0, 2.4, and 3.4-log10 through 75 μs PEF treatment at 13.1, 19.7, and 23.7 kV/cm at 55 °C, respectively (Gurtler and others 2010). Álvarez and others (2003) reported a “ZPEF-value” (that is, a 10-fold increase of the inactivation rate) for PEF inactivation of different Salmonella serovars in citrate–phosphate buffer (pH 7.0) of 15.8 kV/cm (in the range of 19 to 28 kV/cm).
The impact of temperature on the PEF lethality of E. coli in apple juice was systematically studied by Heinz and others (2003) at temperatures ranging from 35 to 70 °C. Energy requirements (which correlates with processing time) to achieve a 7-log10 inactivation of E. coli during PEF processing of apple juice at 24 kV/cm decreased from 160 to 100 kJ/kg when the process temperature was increased from 40 to 50 °C clearly indicating enhanced efficiency of PEF at elevated temperature. E. coli O157:H7 suspended in apple juice is inactivated approximately 0.5, 1.5, and 2.8-log10 by PEF treatments at 20 kV/cm and 20, 30, and 40 °C (initial temperature), respectively (Saldana and others 2011). Similarly, inactivation of Salmonella typhimurium inoculated into citrate–phosphate buffer (pH 3.5) was inactivated 1.5, 2.9, 4.0, and 5.0 log10 CFU/mL after 90 μs PEF treatment at 30 kV/cm and 15, 27, 38, and 50 °C, respectively (Saldana and others 2010). Hence, an increased treatment temperature during PEF processing generally corresponds to increased inactivation of microorganisms at a particular electric field strength.
Finally, pulse shape and pulse width can affect the microbial killing efficiency of PEF. The efficiency of PEF to permeabilize biological cells is largely dependent on the time during which the electric field strength exceeds a certain critical value (Kotnik and others 2003). Therefore, it is generally accepted that exponential decay pulses require more energy than rectangular or square wave pulses since the critical electrical field strength required to kill microorganisms is only exceeded during the first third of the total pulse width. De Haan and others (2002) concluded that exponential decay pulses will not exceed an energy efficiency of 38% compared to square wave pulses at similar peak voltages.
Increasing the pulse width from 0.05 to 3 μs at 50 kV/cm enhanced inactivation of Salmonella enteritidis in 28 mM sodium sulfate with glucose solution (Korolczuk and others 2006). The estimated decimal reduction energy required for PEF destruction of S. enteritidis decreased from 44 to 32 kJ/kg when the pulse width was increased from 0.05 to 1 μs during PEF treatment at 50 kV/cm and 15 °C. Increasing the pulse width from 1 to 3 μs (and taking variations in temperature increase into account) did not further enhance the microbial inactivation efficiency of PEF (Korolczuk and others 2006). Abram and others (2003) examined the effect of pulse width on inactivation of Lactobacillus plantarum in sodium phosphate buffer. Pulse widths of 1, 2, and 5 μs applied at 25 kV/cm were tested for the same treatment times. The results clearly showed that higher levels of L. plantarum inactivation were obtained using a longer pulse width for the same treatment time. On the other hand, increasing the pulse width from 2.5 to 4 μs did not enhance inactivation of L. plantarum in an orange milk beverage during PEF processing at 40 kV/cm (Sampedro and others 2007). These contradictory results can be consequence of temperature variations due to noncontrolled evolution of the temperature during treatments using different PEF systems or pulses. Furthermore, it should be noted that pulse shape, rise and decay time, or polarity have not been found to influence the efficiency of electro-permeabilization of biological cells (Kotnik and others 2003).
Another factor affecting microbial inactivation by PEF is the pH of the treatment medium (Alvarez and others 2006). The low pH of fruit juices acts as a stressor for many bacteria and several studies have investigated its effect on the sensitivity of microorganisms to PEF. For example, E. coli inactivation was drastically increased from 1.7 to 5.7 log10 after PEF treatment at 30 kV/cm and 30 °C for 80 μs when the pH was reduced from 7.0 to 4.0 whereas for Saccharomyces cerevisiae, the pH effect was less pronounced (Aronsson and Rönner 2001). S. typhimurium and E. coli O157:H7 in citrate–phosphate buffer at pH 3.5, 5.2, and 7.0 showed highest resistance to PEF treatment at 30 kV/cm for 99 μs at pH 5.2 independently of the process temperature (4 to 50 °C isothermal conditions) when enumerated straight after the treatment (Saldaña and others 2012). However, another factor to consider is the occurrence of cell damage and microbial survival after PEF treatments. There is evidence that PEF treatment induces sublethally injury in bacterial cells depending on the processing conditions (electrical field strength, temperature, treatment time) and pH of the cell. For example, survival of Salmonella or E. coli at acidic pH (that is, 3.5) was clearly decreased post-PEF treatment at 25 to 35 kV/cm for 100 μs resulting in an additional 1 to 3 log10 reduction of bacteria count within 24-h refrigerated storage (Somolinos and others 2008; Saldaña and others 2010).
PEF inactivation of pathogens in orange juices
Pathogenic strains of Escherichia coli, Salmonella, Staphylococcus, and Listeria monocytogenes continue to cause serious outbreaks of foodborne illness and frequently occur in nonpasteurized orange juices (Ghanshyam and others 2011). The USA federal juice hazard analysis critical control point (HACCP) rule was set up in 2001 after encountering numerous microbiological outbreaks of pathogens in juice. Among other things, the rule stipulates that the juice processors must apply treatments to reduce “pertinent” microorganisms by 99.999% or 5-log10 CFU/mL. The pertinent pathogen in citrus juice is generally regarded as Salmonella (Parish 1998). Thus, for PEF acceptance as an alternative to conventional thermal pasteurization of foodstuffs, regulators often demand the successful destruction of greater than 5-log10 of pathogenic microorganisms in fruit juices.
An overview of published data on PEF inactivation of pathogenic microorganisms in citrus juices is presented in Table 1. The efficacy of PEF against E. coli in fruit juice was demonstrated by Evrendilek and others (1999). A 5-log10 and 5.4-log10 inactivation of E. coli O157:H7 and E. coli 8739, respectively, were achieved in apple juice by applying 29 kV/cm for 172 μs at approximately 25 °C (<35 °C outlet temperature) in a continuous PEF treatment system. Gurtler and others (2010) determined inactivation levels of 1 strain of Enterohemorrhagic E. coli O157:H7 (EHEC), 2 strains of S. typhimurium, and 20 strains of nonpathogenic bacteria in orange juice after PEF treatment at 22 kV/cm for 59 μs at approximately 45 °C (outlet temperature) and at 20 kV/cm for 70 μs and 55 °C (outlet temperature). The level of inactivation for the E. coli O157:H7 strain under each treatment condition was 1.59 and 2.22 log10 CFU/mL, respectively. Nonpathogenic E. coli strain 35218 had a similar level of resistance to PEF treatment under both treatment conditions and is a potential surrogate for pilot plant challenge studies (Gurtler and others 2010).
Table 1. Summary of PEF inactivation of microorganisms in orange juice
|Staphylococcus aureus||3.7||40 kV/cm, 150 μs, 56 °C||5.5||(Walkling-Ribeiro and others 2009b)|
|Listeria innocua||n.d.c||30 kV/cm, 12 μs, 50 °C||6.0||(McDonald and others 2000)|
|Listeria innocua||3.5||40 kV/cm, 100 μs, 56 °C||3.8||(McNamee and others 2010)|
|Escherichia coli||n.d.c||30 kV/cm, 12 μs, 50 °C||6.0||(McDonald and others 2000)|
|Escherichia coli k12||3.5||40 kV/cm, 100 μs, 56 °C||6.3||(McNamee and others 2010)|
|Eschericia coli O157:H7 (EHEC)||3.4||22 kV/cm, 59 μs, 45 °C||1.59||(Gurtler and others 2010)|
| || ||20 kV/cm, 75 μs, 55 °C||2.02|| |
| || ||28 kV/cm, 75 μs, 55 °C||3.79|| |
|Salmonella typhimurium||3.4||22 kV/cm, 59 μs, 45 °C||2.05||(Gurtler and others 2010)|
| || ||20 kV/cm, 70 μs, 55 °C||2.81–3.54|| |
|Samonella typhimurium||n.d.c||90 kV/cm, 50 μs, 50 °C||5.9||(Liang and others 2002)|
|Total aerobic count||4||40 kV/cm, 97 μs, approximately 60 °C||6.2||(Min and others 2003)|
|Yeasts and molds|| || ||5.8|| |
|Aerobic microorganisms, yeasts and moldsb||3.85b||25 kV/cm, 280 μs, T not reported||>3||(Rivas and others 2006)|
|Aerobic plate count||n.d.c||29.5 kV/cm, 60 μs, T not reported||4.2||(Qiu and others 1998)|
|Aerobic plate count||3.78||35 kV/cm, 59 μs, 60 °C||4.0||(Yeom and others 2000b)|
|Yeasts and molds|| || ||4.0|| |
|Zygosaccharomyces bailii||2.9||34.3 kV/cm, 4 μs, 20 °C||3.5 (spores)||(Raso and others 1998b)|
| || || ||5 (veg. cells)|| |
|Saccharomyces cerevisiae (ascospores)||n.d.c||50 kV/cm, approximately 16 μs, 50 °C||2.5||(McDonald and others 2000)|
|Saccharomyces cerevisiae||3.4||12.5 kV/cm, 800 μs, approximately 10 °C||5.8||(Molinari and others 2004)|
|Saccharomyces cerevisiae||3.6||35 kV/cm, 1000 μs, 39 °C||5.1||(Elez-Martinez and others 2004)|
|Pichia fermentans||3.5||40 kV/cm, 100 μs, 56 °C||4.7||(McNamee and others 2010)|
|Byssochlamys fulva (conidiospores)||3.9||34.3 kV/cm, 30 μs, 20 °C||5||(Raso and others 1998a)|
|Neosartoria fischeri (ascospores)||3.9||42.6 kV/cm, 20 μs, 34 °C||<0.1||(Raso and others 1998a)|
|Lactobacillus plantarumb||4.19||35.8 kV/cm., 46.3 μs, T not reported||2.5||(Rodrigo and others 2001)|
|Lactobacillus plantarum||3.4||22 kV/cm, 59 μs, 45 °C||2.57||(Gurtler and others 2010)|
| || ||20 kV/cm, 70 μs, 55 °C||3.07|| |
|Lactobacillus lactis||3.4||22 kV/cm, 59 μs, 45 °C||4.15||(Gurtler and others 2010)|
| || ||20 kV/cm, 70 μs, 55 °C||4.53|| |
|Lactobacillus fermentum||3.4||22 kV/cm, 59 μs, 45 °C||2.11||(Gurtler and others 2010)|
| || ||20 kV/cm, 70 μs, 55 °C||3.22|| |
|Lactobacillus casei||3.4||22 kV/cm, 59 μs, 45 °C||0.43||(Gurtler and others 2010)|
| || ||20 kV/cm, 70 μs, 55 °C||0.60|| |
|Lactobacillus brevis||3.6||25 kV/cm, 150 μs, 32 °C||1.4||(Elez-Martínez and others 2005)|
| || ||35 kV/cm, 1000 μs, 32 °C||5.8|| |
|Leuconostoc mesenteroides||n.d.c||30 kV/cm, 15 μs, 60 °C||5.1||(McDonald and others 2000)|
In the same study, Gurtler and others (2010) detected reductions of 2.8 and 3.5-log10 CFU/mL of S. typhimurium strains UK-1 and 14028 in orange, respectively, after 59 μs treatment at 22 kV/cm and 55 °C. The efficacy of PEF against Salmonella in freshly squeezed orange juice was also demonstrated by Liang and others (2002). PEF treatment in a coaxial chamber with 90 kV/cm for 50 μs at 55 °C, achieved 5.9-log10 inactivation of S. typhimurium in orange juice.
Studies on PEF inactivation of other bacterial pathogens (and their surrogates) in orange juice are limited (Table 1). PEF inactivation kinetics (approximately 10 °C inlet temperature) of Staphylococcus aureus in orange juice indicated D-values of 80, 49.5, and 26.8 μs at 20, 30, and 40 kV/cm, respectively (Walkling-Ribeiro and others 2009b). McDonald and others (2000) reported susceptibility of Listeria innocua in orange juice to PEF processing. Treatments at 30 kV/cm for only 5 μs (42 °C) or 12 μs (50 °C) resulted in 3.5 and 6.0-log10 inactivation of L. innocua, respectively. In contrast, McNamee and others (2010) only found 3.9-log10 inactivation of L. innocua in orange juice after more intense PEF processing at 40 kV/cm and 56 °C for 100 μs. This difference in treatment efficiency might be explained by differences in PEF treatment chamber design, pulse shape (exponential decay compared with rectangular), chemical properties of the orange juice, and variations of bacteria cultivation methods.
L. monocytogenes in apple juice was inactivated by 5 and 6.5 log10 CFU/mL after treatments at 25 kV/cm for 31.5 μs (outlet temperature 50 °C) and 37 μs (outlet temperature 55 °C), respectively. In sour cherry juice, L. monocytogenes was inactivated by 3-log10 CFU/mL upon application of PEF of 27 kV/cm for 131 μs and approximately 20 °C (Altuntas and others 2011).
Shelf-life extension of orange juices by PEF
PEF is effective at inactivating spoilage microorganisms in liquid foods such as juices thereby extending shelf life and often improving quality (Min and others 2007). A summary of selected papers on microbial inactivation and shelf-life extension of orange and other juices by PEF is provided in Table 1.
PEF inactivation of spoilage microorganisms in orange juice was demonstrated by Timmermans and others (2011) where PEF treatment (23 kV/cm, 36 μs, inlet temperature 38 °C, outlet temperature 58 °C) resulted in microbial counts at levels less than the detection limit for up to 2 mo of refrigerated (4 °C) storage. At the end of the shelf life (58 d at 4 °C), PEF-treated orange juice exhibited microbial counts of <10 CFU/mL for Enterobacteriaceae and yeast and mold, <100 CFU/mL for lactic acid bacteria and, <1000 CFU/mL for total plate count (that is, all counts were less than the respective detection limits). Hence, the study demonstrated that PEF is an effective process to increase the microbial shelf life of orange juice from 9 (fresh juice) to up to 2 mo when stored at 4 °C.
Orange juice treated at 29.5 kV/cm for 60 μs at 30 °C inlet temperature and aseptically bottled was shelf stable at 4 °C for 7 mo whereas untreated juice spoiled (that is, total plate count >106 CFU/mL) after 30 d (Qiu and others 1998).
PEF was employed in inactivating spoilage microorganisms in juice-milk blends. The combined effect of thermal and PEF processing on the shelf life of an orange juice-milk beverage (OJMB) was studied by Sampedro and others (2009). Thermal treatment conditions were 85 °C for 66 s whereas PEF treatment conditions were 30 kV/cm for 50 μs, inlet temperature 65 °C, and outlet temperature 80 °C (estimated 5 s until cooling). The reductions in bacterial as well as yeast and mold counts were similar after PEF or thermal treatments (4.5 and 4.1-log10 CFU/mL for thermal compared with 4.5 and 5.0-log10 CFU/mL for PEF). The estimated shelf lives were 2 and 2.5 wk at 8 to 10 °C for the thermally and PEF processed OJMB. Therefore, PEF-treated OJMB achieved a slightly higher shelf life than the thermally treated OJMB. For an OJMB, treatment with PEF achieved the equivalent enzyme inactivation as the thermal treatment, but better preserved color and flavor compounds (Sampedro and others 2009). However, it should be noted that thermal processing times (66 s) used in this study are slightly greater than typically used in industry and should have yielded better microbial kill and stability.
A study by Sharma and others (1998) produced a PEF-treated protein-fortified orange juice with a microbiologically stable shelf life of 5 mo at 4 °C. Jia and others (1999) achieved a 6-wk shelf life at 4 °C for PEF-treated (30 kV/cm for 240 μs at 25 °C initial temperature) orange juice. This PEF treatment resulted in a total aerobic plate count and yeast and mold count below the detection limit (<1 CFU/mL).
Yeom and others (2000b) achieved a 112-d shelf life at 4 °C for orange juice treated by PEF. The PEF treatment applied was 35 kV/cm for 59 μs at 25 (inlet) to 59 °C (outlet) temperature and the inactivation was comparable to a thermal treatment at 94.6 °C for 30 s. The PEF-treated orange juice maintained low levels of microorganisms (approximately 1 log10 total plate count) indicating that the PEF treatment resulted in irreversible damage of cells.
Min and others (2003) demonstrated that commercial-scale (500 L/h) PEF treatment of orange juice at 40 kV/cm for 97 μs at 45 to 65 °C (inlet compared with outlet temperature) reduced the total aerobic plate count and yeast and mold count by 6 log10 CFU/mL, achieved a shelf life of 196 d at 4 °C.
PEF was also successfully applied for refrigerated shelf-life extension of apple cider, cranberry juice, and fruit smoothies. For example, Evrendilek and others (1999) reported an increase of more than 67 d of microbial shelf life for apple cider stored at both 4 °C and 22 °C following a PEF treatment at 35kV/cm for 94 μs and a thermal treatment of 60 °C for 30 s. Jin and Zhang (1999) demonstrated more than 4-log10 inactivation of total aerobic bacteria, yeasts, and molds in cranberry juice with a PEF treatment of 35 kV/cm for 195 μs (at 15 to 25 °C Tmax). The spoilage microorganisms were inhibited and the treated and aseptically packaged cranberry juice had a shelf life of 8 mo, 37 d and 30 d at 4, 22 and 37 °C, respectively.
A combination of mild heat (55 °C for 60 s) and PEF treatment (34 kV/cm for 60 μs at 55 °C [outlet]) successfully extended the refrigerated (4 °C) shelf life of a tropical fruit juice smoothie (50% pineapple, 28% banana, 12% apple, 7% coconut, 3% orange juice) to 21 d (Walkling-Ribeiro and others 2010). This treatment reduced the total plate count of the smoothie by approximately 4.4-log10 which was similar to microbial reductions found after 15 s thermal processing at 72 °C. Interestingly, the thermally treated fruit smoothie exhibited only 14 d refrigerated shelf life and had similar sensory quality to the smoothie treated under combined mild heat and PEF.
Lactic acid bacteria are the predominant microorganisms, in spoiled orange juice and L. plantarum is one of the most common species. Gómez and others (2005) developed a model describing PEF inactivation of L. plantarum in McIlvaine buffer of varying pH (3.5 to 7.0) and tested the validity for PEF-treated orange and apple juices in a batch PEF system. L. plantarum was more sensitive to PEF at higher electric field strengths and in media of low pH. For example, treatment at 22 kV/cm for 400 μs (temperature was kept at <35 °C) inactivated around 5.0, 2.8, 2.6, and 1.1-log10 CFU/mL of L. plantarum at pH 3.5, 5.0, 6.5, and 7.0, respectively. A Weibull-type model accurately described upward concave survival curves of L. plantarum. The model satisfactorily predicted the inactivation of L. plantarum in apple and orange juices but is not suitable for commercial application of PEF processing due to the relatively long treatment time (400 μs) used. A comparative study, which included a range of bacterial species, indicated that a strain of Lactobacillus casei had the greatest resistance to PEF treatments of all the strains tested in orange juice, possibly because of its comparatively high heat and acid tolerance (Gurtler and others 2010). Treatment of L. casei in orange juice at 20 kV/cm for 70 μs and 55 °C outlet temperature achieved an inactivation level of 0.6-log10 CFU/mL, compared with levels of inactivation for L. plantarum, L. fermentum, and L. lactis of 3.07, 3.22, and 4.75-log10, respectively (Table 1).
Lactobacillus brevis was inactivated to 5.8 log10 CFU/mL in orange juice during PEF processing at 35 kV/cm for 1000 μs at less than 32 °C (4-μs pulse width in bipolar mode) (Elez-Martínez and others 2005). However, this very long treatment time for PEF processing is uneconomical as it is accompanied by very high consumption of electrical energy (several MJ/L).
Yeasts are also major spoilage microorganisms of orange juice. Yeasts (that is, S. cerevisiae) are relatively PEF sensitive and inactivation requires much lower electrical field intensity and/or pulsing time than inactivation of E. coli or L. innocua (Aronsson and others 2005). Elez-Martinez and others (2004) achieved a 5.1-log10 inactivation of S. cerevisiae suspended in orange juice after exposure to 1000 μs PEF treatment at 35 kV/cm and temperatures lower than 39 °C. However, similar inactivation levels of S. cerevisiae in orange juice were also obtained at much lower PEF intensities (12.5 kV/cm, 800 μs, 10 °C) (Molinari and others 2004). Ascospores of S. cerevisiae are less susceptible to PEF treatments than the vegetative cells. Nonetheless, a 2.5-log10 reduction in orange juice was achieved after treatment at 50 kV/cm and 50 °C for 16 μs (McDonald and others 2000). Similarly, PEF treatment for 4.0 μs at 34.3 kV/cm and 20 °C maximum temperature inactivated 4.7-log10 CFU/mL of vegetative cells but only 3.8-log10 CFU/mL of ascospores of Zygosaccharomyces bailii in orange juice (Raso and others 1998b).
PEF inactivation of microorganisms in combination with secondary hurdles
PEF inactivation of microorganisms can be enhanced by the addition of antimicrobials such as nisin. Using PEF conditions of 80 kV/cm at 20 pulses, and a temperature of 44 °C, with 100 U nisin/mL resulted in greater than 6-log10 reduction in total plate count in orange juice (pH 3.5) compared to only 1.75-log10 inactivation of spoilage organisms in the absence of nisin (Hodgins and others 2002). The antimicrobial mechanism acts on the bacteria cell membrane, creating a greater electroporation effect and thus leading to cell death.
Other hurdles such as the use of benzoate, sorbate, citric acid (CA) enhance the performance of PEF. Benzoate, sorbate, and CA were coupled with PEF treatment to inactivate E. coli O157:H7 (ATCC 43895) and a nonpathogenic E. coli (ATCC 35218) in strawberry juice Gurtler and others (2011). The 2 strains were inoculated into single-strength strawberry juice with or without 750 ppm sodium benzoate (SB), 350 ppm potassium sorbate (PS), and 2.7% CA. The juice was treated at 18.6 kV/cm for 150 μs and outlet temperatures of 45, 50, and 55 °C. Inactivation levels of surrogate E. coli (ATCC 35218) at 45, 50, and 55 °C were 2.86, 3.12, and 3.79 log10 CFU/mL, respectively, in control juice (pH 3.4), and 2.75, 3.52, and 5.11 with the addition of benzoate, sorbate, and CA (pH 3.5). Inactivation of E. coli O157:H7 under equivalent conditions were 3.09, 4.08, and 4.71-log10 CFU/mL, respectively, and 2.27, 3.29, and 5.40-log10 with antimicrobial. E. coli O157:H7 in strawberry juice with antimicrobials and 2.7% CA (pH 2.7) treated with PEF was inactivated 2.60, 4.32, and 6.95-log10 CFU/mL at 45, 50, and 55 °C while the surrogate E. coli decreased by 3.54, 5.69, and 7.13 log10 under equivalent conditions. This indicates that E. coli ATCC 35218 provides a margin of safety when used as a PEF surrogate for O157:H7 in control strawberry juice or in strawberry juice with the addition of selected antimicrobials.
PEF Inactivation of Native Enzymes in Citrus Juices
PEF exhibits limited effects on enzymes, perhaps due to conflicting reports in the literature on the thermal effects associated with PEF, which is often credited for the observed enzyme inactivation (Terefe and others 2013). Nevertheless, several research studies concluded that under selected processing conditions, PEF results in substantial inactivation of many indigenous food enzymes (Giner and others 2005; Espachs-Barroso and others 2006; Min and others 2007; Buckow and others 2012). Conformational changes in enzymes following PEF treatment were observed by Yeom and others (1999). However, compared to microorganisms, more intense PEF treatments are required to inactivate enzymes (Yeom and others 2002). On the other hand, mild PEF treatments are reported to enhance the activity of some enzymes such as fungal polygalacturonase (Giner and others 2003).
The mechanism of PEF inactivation of enzymes is not well understood. The available scientific literature suggests that both electrochemical and thermal effects associated with PEF individually or in synergy result in changes in the structure and conformation of enzymes, which may lead to inactivation (Terefe and others 2013). The structure of proteins is stabilized by a sensitive balance of various covalent peptide bonds and noncovalent interactions such as hydrogen bonds and hydrophobic, electrostatic and Van der Waals interactions. The application of an external electric field may affect the local electrostatic fields in proteins and disrupt electrostatic interactions of peptide chains leading to conformational changes. In addition, PEF-induced electrolysis and free radical formation may result in localized (that is, near electrodes) pH shifts in watery systems, and oxidation of amino acid residues important for the activity and stability of enzymes (Meneses and others 2011). The instantaneous temperature increase during pulsing as well as localized hot spots in the PEF treatment chamber may also contribute to PEF-induced denaturation of enzymes (Schilling and others 2008). A synergy between heat and PEF reported in a number of studies is attributed to increased mobility of charged groups at higher temperatures that affects electrostatic interactions and the stability of proteins (Yang and others 2004).
Orange pectin methylesterase (PME) is partially inactivated by PEF processing at elevated temperatures (Table 2). Yeom and others (2000b) reported approximately 90% inactivation of PME in orange juice after application of 35 kV/cm for a total treatment time of 59 μs with an outlet temperature of approximately 60 °C. In a subsequent study, 90% PME inactivation in orange juice was achieved by application of 25 kV/cm for 2 μs pulses for 250 μs at a maximum outlet temperature of 64 °C (Yeom and others 2002). Sentandreu and others (2006) reported more than 90% inactivation of orange juice PME after PEF processing at 25 kV/cm using 2 μs, square wave bipolar pulses for up to 330 μs and 72 °C outlet temperature.
Table 2. Summary of representative studies on PEF inactivation of pectin methylesterase (PME) in citrus juices
|Orange juice||35 kV/cm, 59 μs, 60.1 °C||90%||Pulse width used was 1.4 μs||(Yeom and others 2000b)|
|Orange juice||25 kV/cm, 250 μs, approximately 64 °C||90%||2-μs pulse width||(Yeom and others 2002)|
|Orange juice||80 kV/cm, 60 μs, 44 °C||92.7% (estimate)||Exponential pulsed of 2 to 3 μs applied in a batch system||(Hodgins and others 2002)|
|Orange juice||20 kV/cm, 4000 μs, approximately 25 °C||≤10%||Increased PME activity in some cases||(Van Loey and others 2002)|
| ||30 kV/cm, 400 μs, approximately 25 °C|| || || |
|Orange juice||25 kV/cm, 330 μs, 72 °C||90.6%||Continuous treatment with 2-μs rectangular pulses applied in bipolar mode. Note that 1.5 s time needed for the juice to reach the cooling.||(Sentandreu and others 2006)|
|Orange juice||17.5 kV/cm, 400 μs, 67.2 °C||87%||Note that 40-μs rectangular pulses applied in monopolar mode in a batch system. Thermal effects of PME inactivation during PEF treatments are subtracted.||(Espachs-Barroso and others 2006)|
|Orange juice||35 kV/cm, 1500 μs, 37.5 °C||78.1%||Continuous treatment with 4-μs rectangular pulses applied in bipolar mode. Estimated energy input was 8.085 MJ/L||(Elez-Martínez and others 2007)|
|Orange juice||30 kV/cm, 50 μs, approximately 46 °C||20%||Continuous treatment in a parallel plate system with 1-μs rectangular pulses in monopolar mode||(Walkling-Ribeiro and others 2009b)|
|Orange juice||23 kV/cm, 36 μs, 58 °C||Approximately 40%||Continuous treatment with 2-μs rectangular pulses in monopolar mode. Estimated energy input was 76 kJ/L||(Vervoort and others 2011)|
|Clementine juice||25 kV/cm, 330 μs, 72 °C||88.3%||Continuous treatment with 2-μs rectangular pulses applied in bipolar mode. Note that 1.5 s time needed for the juice to reach the cooling.||(Sentandreu and others 2006)|
|Grapefruit juice||20 kV/cm, 25 μs, approximately 25 °C||59%||Continuous treatment in a parallel plate system with 1-μs rectangular pulses in monopolar mode||(Riener and others 2009)|
| ||30 kV/cm, 100 μs, 60 °C||90%|| || |
|Orange–carrot blend (4:1 v/v)||25 kV/cm, 340μs, 63°C||81.4%||Continuous treatment with 2.5-μs rectangular pulses applied in bipolar mode||(Rodrigo and others 2003)|
|Orange–carrot blend (4:1 v/v)||25 kV/cm, 330μs, 70°C||81%||Continuous treatment with 2.5-μs rectangular pulses applied in bipolar mode||(Rivas and others 2006)|
|Orange–carrot blend (1:1 v/v)||24 kV/cm, 89 μs, 49°C||14%||Monopolar pulses with a pulse width of 1μs||(Caminiti and others 2012)|
Many of the PEF studies report substantial inactivation (78 to 92.7%) of PME in orange and other citrus juices following PEF processing at selected conditions (see Table 2). Nevertheless, very high-specific energy inputs were used in many of these studies to achieve a substantial level of PME inactivation, greater energy than is typically required to inactivate vegetative microorganisms by PEF, economical for commercial scale PEF processing. The treatment times proposed are also much longer than the commercial practical limits (<200 μs). Interestingly, substantial inactivation (up to 62%) of the enzyme in red grape juice was already achieved after only 25 μs at 30 kV/cm and 20 °C (inlet temperature) (Riener and others 2009).
Studies on the effect of PEF on the thermo-stable fraction of orange PME and or results that achieved >97% inactivation of the enzyme in orange juice were not identified. Thus, the thermo-stable fraction approximately 10% of the total orange PME activity is resistant as well, since no complete inactivation of PME is reported. Nonetheless, PEF-treated citrus juices featured good cloud stability during storage comparable to thermally pasteurized juices (Yeom and others 2000a; Hodgins and others 2002).
Orange peroxidase (POD) was susceptible to inactivation by PEF with intense long treatment times. Elez-Martínez and others (2006) reported 93% and 100% inactivation of POD in orange juice after treatment at 35 kV/cm for 1000 and 1500 μs (at less than 40 °C), respectively. Bipolar pulses of 4 μs width and at a pulse frequency of 200 Hz were used in the study. According to the authors, the temperature was maintained below 35 °C despite the long treatment time. Vervoort and others (2011) reported a 30% reduction of orange POD upon application of 2 μs pulses at 23 kV/cm and 35 to 58 °C for much shorter total treatment times of 36 μs. To our knowledge, there are no further reports on PEF inactivation of oxidases in orange juice.
Quality Aspects of PEF-Treated Orange Juice during Shelf Life
PEF processing exhibits potential to effectively inactivate vegetative microorganisms in orange juice at moderate temperatures while only minimally impacting nutritional and sensorial quality (Min and others 2007). Usually, PEF treatment at room to moderate temperature (<50 °C) exhibits limited, detrimental effect on the sensorial and nutritional quality of orange juice (Min and others 2003; Sánchez-Moreno and others 2005; Cserhalmi and others 2006; Rivas and others 2007). PEF processing performed at ≤68 °C resulted in less degradation of vitamin C (Min and others 2003; Torregrosa and others 2006; Elez-Martínez and Martín-Belloso 2007), carotenoids (Cortés and others 2006; Torregrosa and others 2006), polyphenols (Sánchez-Moreno and others 2005; Sánchez-Moreno and others 2009), and volatile aroma compounds (Qiu and others 1998; Hodgins and others 2002) in orange juice compared to thermal pasteurization (for example, 95 °C for 30 s). Moreover, substantially greater retention of ascorbic acid (Qiu and others 1998; Min and others 2003; Elez-Martínez and Martín-Belloso 2007), carotenoids (Cortés and others 2006), and volatile content and better sensory profile (Jia and others 1999; Yeom and others 2000a) was observed in PEF-treated (35 kV/cm, 59 μs, approximately 60 °C) orange juices after processing and during refrigerated storage compared to thermally treated (95 °C for 30 s) orange juice. PEF-treated orange juice exhibited less browning, greater whiteness (L*) values, and a smaller particle size distribution than heat-pasteurized orange juice during storage at 4 °C. °Brix, pH, and viscosity of orange juice or a fruit smoothie were unaffected by PEF processing and, indeed, these quality parameters were better retained during refrigerated storage than in heat-pasteurized juices and smoothies (Min and others 2003; Walkling-Ribeiro and others 2010). Furthermore, none-enzymatic browning, hydroxymethylfurfurol, and organic acid content were not significantly (P ≤ 0.05) changed by PEF-treatments (see Table 3). Nevertheless, overall there is little evidence that PEF-treated orange juice significantly differs in refrigerated shelf life, sensorial quality, and consumer acceptability from heat-pasteurized juices (Walkling-Ribeiro and others 2009a). It is also worth noting that thermally treated orange juice (for example, 98 °C for 11 s) often does not show significant flavor changes when compare to fresh orange juice and the loss of vitamin C during storage is significantly greater than the one caused by processing (Moshonas and Shaw 1997; Min and others 2007).
Table 3. Summary of representative studies on the effect of PEF on selected vitamins, phytophenols, and volatiles in orange juice
|Vitamin C||35 kV/cm, 59 μs, approximately 60 °C||Unchanged||Approximately 75% retention after 110 d in PETb at 4 °C||(Ayhan and others 2001)|
|Vitamin C||40 kV/cm, 97 μs, approximately 65 °C||Unchanged||Approximately 50% retention after 50 d in PEc cups at 4 °C||(Min and others 2003)|
|Vitamin C||35 kV/cm, 59 μs, approximately 60 °C||Approximately 1%||Approximately 50% retention after 40 d in PEc cups at 4 °C||(Yeom and others 2000a)|
| || || ||Approximately 10% retention after 12 d in PEc cups at 22 °C|| |
|Vitamin C||25 kV/cm, 110 μs, approximately 50 to 60 °C||Approximately 7%||n.d.d||(Torregrosa and others 2006)|
|Vitamin C||25 kV/cm, 400 μs, approximately 65 °C||10%||n.d. d||(Elez-Martínez and Martín-Belloso 2007)|
|Vitamin C||28 kV/cm, 100 μs, approximately 34 °C||Approximately 3%||n.d.d||(Cserhalmi and others 2006)|
|Vitamin C||30 kV/cm, 50 μs, approximately 46 °C||6%||n.d.d||(Walkling-Ribeiro and others 2009b)|
|Vitamin A||30 kV/cm, 100 μs, approximately 40 °C||Approximately 8%||Approximately 82% retention after 6 wk in Elopack packages at 10 °C||(Cortés and others 2006)|
|Vitamin A||35 kV/cm, 750 μs, approximately 50 °C||Approximately 8%||Unchanged after 40 d at 4 °C in opaque tubes with nitrogen headspace||(Plaza and others 2011)|
|Citric acid||28 kV/cm, 100 μs, approximately 34 °C||Approximately 1%||n.d.d||(Cserhalmi and others 2006)|
|Citric acid||28 kV/cm, 100 μs, 27 °C||Approximately 1%||n.d.d||(Hartyáni and others 2011)|
|Total carotenoids||30 kV/cm, 100 μs, approximately 40 °C||Approximately 8%||Approximately 90% retention after 6 wk in Elopack packages at 10 °C||(Cortés and others 2006)|
|β-carotene||35 kV/cm, 750 μs, approximately 50 °C||Approximately 1%||Unchanged after 40 d at 4 °C in opaque tubes with nitrogen headspace||(Plaza and others 2011)|
|Flavonoids||35 kV/cm, 750 μs, approximately 50 °C||Unchanged||n.d.d||(Sánchez-Moreno and others 2005)|
|Total flavanone||35 kV/cm, 750 μs, approximately 50 °C||Approximately 15% increase||Approximately 50% retention after 40 d at 4 °C in opaque tubes with nitrogen headspace||(Plaza and others 2011)|
|Hydroxymethylfurfural||25 kV/cm, 280 μs, approximately 68 °C||Unchanged||No change after 8 wk at 2 °C||(Rivas and others 2007)|
|Hydroxymethylfurfurol||28 kV/cm, 100 μs, approximately 34 °C||Approximately 10%||n.d.d||(Cserhalmi and others 2006)|
|α-Pinene||35 kV/cm, 59 μs, approximately 60 °C||n.d. d||Approximately 45% retention after 112 d in PEc cups at 4 °C||(Yeom and others 2000a)|
|Myrcene||35 kV/cm, 59 μs, approximately 60 °C||n.d. d||Approximately 50% retention after 56 d in PEc cups at 4 °C||(Yeom and others 2000a)|
|Myrcene||40 kV/cm, 97 μs, approximately 65 °C||10%||Approximately 50% retention after 70 d in PETb at 4 °C||(Min and others 2003)|
|β-mircene||28 kV/cm, 100 μs, 27 °C||16% increase||Glass bottles at 4 °C||(Hartyáni and others 2011)|
|δ-Limonene||35 kV/cm, 59 μs, approximately 60 °C||Unchanged||Approximately 87% retention after 110 d in PEc cups at 4 °C||(Ayhan and others 2001)|
|Limonene||28 kV/cm, 100 μs, 27 °C||22% increase||n.d.d||(Hartyáni and others 2011)|
|Linalool||28 kV/cm, 100 μs, approximately 34 °C||Unchanged||n.d.d||(Cserhalmi and others 2006)|
|Linalool||28 kV/cm, 100 μs, 27 °C||Approximately 4%||n.d.d||(Hartyáni and others 2011)|
|Octanal||35 kV/cm, 59 μs, approximately 60 °C||n.d. d||Approximately 59% retention after 112 d in PEc cups at 4 °C||(Yeom and others 2000a)|
|Decanal||35 kV/cm, 59 μs, approximately 60 °C||Approximately 5%||Approximately 50% retention after 30 d in PEc cups at 4 °C||(Yeom and others 2000a)|
|Decanal||28 kV/cm, 100 μs, approximately 34 °C||Unchanged||n.d.d||(Cserhalmi and others 2006)|
|Decanal||28 kV/cm, 100 μs, 27 °C||45%||n.d.d||(Hartyáni and others 2011)|
|Valencene||28 kV/cm, 100 μs, approximately 34 °C||Unchanged||n.d.d||(Cserhalmi and others 2006)|
|Valencene||28 kV/cm, 100 μs, 27 °C||Approximately 5% increase||n.d.d||(Hartyáni and others 2011)|
|Naringenin||35 kV/cm, 750 μs, approximately 50 °C||Approximately 12%||Approximately 67% retention after 40 d at 4 °C in opaque tubes with nitrogen headspace||(Plaza and others 2011)|
|Hesperetin||35 kV/cm, 750 μs, approximately 50 °C||Approximately 3%||Approximately 52% retention after 40 d at 4 °C in opaque tubes with nitrogen headspace||(Plaza and others 2011)|
PEF Efficacy at Elevated Temperature
Even though PEF processing is considered a nonthermal technology, temperature is an important processing parameter. Joule heating effects result in temperature increases during PEF processing and can play a significant role in PEF inactivation of microorganisms and enzymes as well as degradation of heat labile bioactives.
Elevating the treatment temperature from room to between 35 and 60 °C can make use of the synergetic effects of mild heat on the PEF treatment efficiency to inactivate E. coli in fruit juice (Heinz and others 2003; Toepfl and others 2007). Using moderate temperatures around 30 °C (Aronsson and Rönner 2001) and 45 to 60 °C (Jayaram and others 1992) enhanced the inactivation of microorganisms by PEF. The influence of temperature on microbial inactivation by PEF was demonstrated by Shin and Pyun (1999), who treated L. plantarum with exponential pulses at a selected field strength of 80 kV/cm applied for 1000 μs in phosphate buffer and achieved inactivation levels of approximately 2.3 and 5.2-log10 at 30 and 50 °C, respectively. Temperature can also be used to enhance microbial inactivation at a constant energy input (Wouters and others 2001). Application of moderate heat (45 and 55 °C) and PEF in a sequenced hurdle approach also showed synergistic effects on inactivation of E. coli in a fruit smoothie (Walkling-Ribeiro and others 2008). The influence of temperature on microbial inactivation by PEF is probably based on changes to the cell membrane with an increase in temperature. At temperatures lower than approximately 30 °C, the phospholipid bilayer of the cell membrane is closely packed and exhibits a gel-like structure. As temperatures increase, the structure becomes less ordered, loses its elastic properties, and transforms into a liquid crystalline state (Aronsson and Rönner 2001). This is combined with a reduction of the bilayer thickness, making the cell more vulnerable to pore formation (Jayaram and others 1992). Determination of the exact treatment temperature is often difficult, so it is important to control the temperature and report inlet and outlet temperatures. Due to ohmic heating, it is not possible to avoid an increase in temperature during PEF treatments; therefore, intermediate cooling is necessary to minimize exposure of the fruit juice to high temperatures.
Increasing the process temperature can also enhance the inactivation of enzymes during PEF processing (Terefe and others 2013). In fact, most enzymes, including orange PME, are stable during PEF processing near room temperature and excessively large specific energy inputs and/or treatment times are required to incur substantial inactivation. For example, a specific energy input of 44 MJ/L and 8000 μs treatment time were used to achieve 93.8% inactivation of tomato PME at the reported maximum temperature of 15 °C (Giner and others 2000). Similarly, Elez-Martínez and others (2007) applied 35 kV/cm for 1500 μs (approximately 8 MJ/L) at approximately 35 °C to achieve 77% inactivation of PME in orange juice. Application of such high-specific energy also requires excessive cooling energy (in the same order of magnitude), which makes these results impractical (and most probably uneconomical) to be applied at commercial scale.
Not many studies have investigated the effect of temperature on PEF inactivation of enzymes in a systematic manner. Yeom and others (2002) reported approximately 90% inactivation of orange PME by PEF treatment of 25 kV/cm for 250 ms in a 50 °C (Tmax approximately 66 °C) water bath and only 36.3% inactivation when an equivalent PEF treatment was carried out in a 10 to 20 °C water bath (Tmax 48 °C), indicating synergistic inactivation effects of mild heat and PEF. Similarly, Schilling and others (2008) observed limited inactivation of apple oxidases when the inlet temperature was 20 or 40 °C. Increasing the inlet temperature to 60 °C resulted in complete inactivation of oxidative enzymes at an electric field strength of 35 kV/cm and 63.4 kJ/kg specific energy input. The contribution of heat to PEF inactivation of apple POD and polyphenol oxidase was estimated at about 33.3% (Schilling and others 2008), clearly indicating synergistic effects of mild heat and PEF. The synergistic effect of mild heat and PEF is often attributed to an increased mobility of charged groups at high temperatures, which affects electrostatic interactions and the stability of proteins including enzymes (Yang and others 2004).
State of Commercialization and Industrial PEF Experience
Although many potential applications of PEF were identified for the food industry (Jeyamkondan and others 1999; Huang and Wang 2009), most of the current commercial applications of PEF are in the area of fruit juice preservation and pretreatments of vegetables and tubers before cutting, blanching, or extraction of valuable or undesired compounds. Most of the orange juice consumed in Europe is imported as concentrate from Brazil or the United States and often exhibits poor sensorial quality attributes after reconstitution. Thus, in contrast to the United States, a market shift from shelf stable toward fresh, often unpasteurized fruit juices with short chill-stability is occurring in Europe as well as other parts of the world, impacting the fruit juice supply chain (AIJN 2012). Fruit juice beverages are no longer produced from concentrates but manufactured from fresh or frozen fruits close to the point of sale. The chill-stability of such products is often in the range of 7 to 10 d depending on raw material quality and production processes. To supply a large variety of beverages and allow sufficient time on the market shelf, significant distribution efforts and frequent collection and distribution of dated beverages is required.
In the 1990s, consortia of food processors, equipment manufacturers, and universities were formed in the United States and Europe to develop PEF applications and equipment (Toepfl and others 2006). In 1995, a continuous PEF processing system was launched by PurePulse, a subsidiary of Maxwell Laboratories. In 2006, a 1st commercial PEF installation for organic fruit juice preservation was approved by the US Food and Drug Administration (FDA) (Clark 2006). However this application was aborted in 2008 due to technical and commercial limitations.
In Europe, the first commercial PEF operation was started in 2009 with the installation of a 1500 L/h juice preservation line (Toepfl 2012). The PEF treatment was performed at an initial temperature of 40 °C, a field strength of 20 kV/cm, and an energy input of approximately 120 kJ/kg resulting in a final temperature of less than 60 °C (using intermediate cooling) and 5-log10 reduction of the total bacteria count in orange juice (Toepfl 2011). In addition to inactivation of natural microbial flora, the inactivation of inoculated E. coli 35218 was also demonstrated. A 5-log10 inactivation of inoculated E. coli 35218 is a key requirement for US FDA approval. When operating at acceptable maximum temperatures in a range of 50 °C the orange juice shelf life is extended to approximately 21 d, while maintaining fresh-like taste and quality. At present PEF-treated fruit juices and smoothies are on market shelves in Germany, the Netherlands, and the United Kingdom, utilizing PEF processing equipment with a capacity of 1500 to 2000 L/h and 5000 to 8000 L/h (Irving 2012). To reduce the extent of spoilage and quality degradation due to enzyme reactions, a chilled distribution chain is required. Although successful in a niche market of fresh juices with short refrigerated shelf life, PEF has not yet been adapted by a major orange juice processor, possibly because of the investment challenges for processing plants that already have a large capacity for thermal pasteurization and the way orange juice and concentrates are currently distributed globally (either frozen or nonrefrigerated).
The total processing costs, including investment and operation, are in a range of 0.02 to 0.03 USD/L of product (Toepfl and others 2006). This is justified by an extended chill-stability allowing wide distribution and market range as well as drastically reducing product return. The impact of PEF on fruit juice quality parameters as well as the level of undesired substances were evaluated with regard to the European Novel Food Legislation (Cserhalmi and others 2006; Schilling and others 2008; Timmermans and others 2011; Vervoort and others 2011). Similar to other new food preservation processes such as high-pressure processing, no significant (P ≤ 0.05) process-induced changes were observed in PEF-treated fruit juices and, therefore, the application of PEF was not considered “novel.”
A 3 to 4-log10 inactivation of the total microbial count was observed in fresh carrot juice after a PEF treatment at 30 °C and an energy input of 100 kJ/kg at a scale of 500 L/h. The chill shelf life of carrot juice was extended from 2 to 6 d despite the low acid pH. Vegetable juices are of great importance in Asia and Eastern Europe, due to a large content of nutrients and health-related substances and PEF will continue to be of great interest as an alternative processing technique.
Whenever applied in food industry, the use of new processes also requires suitable process control options and adoption of HACCP and hazard and operability study concepts. PEF preservation is a HACCP critical control point which requires consistent delivery of sufficient treatment intensity and record keeping at all times (Nöhle 1994). In general, HACCP concepts similar to the concepts for thermal processing can be applied for PEF processing of foodstuff. In a 1st step, product characteristics (that is, pH, electrical conductivity, composition, and so on) must be considered in the design of the process (Vega-Mercado and others 1996). The required energy delivery, optimal pulse width, pulse shape, and frequency at the desired treatment temperature need to be elaborated in laboratory-scale challenge tests for each food product. At industrial-scale, the required energy delivery is often ensured by continuous monitoring of power delivery, validated by measurement of the food's temperature increase. Furthermore, treatment chamber design and geometry as well as electrical components of the system are critical elements ensuring safety of both equipment operation and product integrity (Vega-Mercado and others 1996; Buckow and others 2011).
Several studies demonstrate the effectiveness of PEF processing for microbial inactivation in fruit juices including orange juice, and subsequent shelf-life extension. Generally, yeasts and molds are relatively sensitive to PEF processing and are effectively inactivated in orange juice by application of mild PEF conditions (that is, <10 kV/cm for <20 μs) near room temperature. Critical and relatively PEF resistant microorganisms in orange juice are lactic acid bacteria and pathogenic E. coli. These microorganisms exhibit resistance to PEF under economically feasible PEF treatment conditions near room temperature. Nevertheless, elevating the PEF processing temperature from room to approximately 50 °C enhances PEF process efficiency, reduces treatment time, and reduces specific energy requirements to achieve a 5-log10 reduction of E. coli in fruit juice. Furthermore, PEF inactivation of microorganisms is enhanced by the addition of antimicrobials such as nisin, benzoate, sorbate, and CA, allowing the energy-efficient use of PEF for fruit juice preservation.
PEF exhibits limited effects on enzymes at room to moderate temperatures. Nevertheless, several studies reported inactivation of orange PME and POD at specific PEF conditions. However, compared to microbial inactivation more intense PEF treatments and higher process temperatures are required to inactivate enzymes. It is estimated that 4 to 5 times the specific energy input is needed for enzyme inactivation by PEF compared to thermal processing. PEF inactivation of orange PME is considered to be due to the combined effect of PEF and temperature. No studies on the effect of PEF on the thermostable fraction of orange PME (approximately 10% of total PME activity) and/or results that achieved >97% inactivation of the enzyme in orange juice were identified. Nonetheless, PEF-treated orange juices exhibited cloud stability during storage comparable to thermally pasteurized juices produced at pilot scale.
PEF treatment at room to moderate temperatures results in limited detrimental effects on the sensorial and nutritional quality of orange juice. In fact, PEF processing under conditions suitable to ensure microbial safety and stability can result in less degradation of vitamin C, carotenoids, polyphenols, and volatile aroma compounds in orange juice than conventional thermal pasteurization (for example, 95 °C for 30 s). However, possibly this could also be achieved by good deaeration before heat treatment of orange juices. Other physicochemical aspects of orange juice such as color, particle size, turbidity, or viscosity are either enhanced or equivalent to thermally processed orange juices after PEF processing at elevated temperatures. Hence, there is little evidence that orange juice treated at suitable PEF conditions significantly differs in shelf life during refrigerated storage from heat-pasteurized juices but may exhibit improved aroma, volatile, vitamin C retention, and overall sensory profile.
Although now established in northern European countries where no oranges can be grown, perhaps one of the greatest opportunity for the production PEF-pasteurized orange juice is in orange producing regions with high-disposable incomes such as Mediterranean countries, Florida, or Australia. Here, the economic challenge may arise from the fact that much of the oranges grown in these regions are destined to the fresh market and the costly agricultural practices associated with it.
The recent development of pulsed power systems suitable for food industry applications is promoting successful transfer of scientific knowledge on PEF effects on microorganisms, enzymes, and sensory properties of fruit juices from laboratory scale to industrial scale (up to 8000 L/h). New applications and foods presented with PEF technology are likely to emerge in areas where thermal processing results in major damage to the food, including citrus juice blends with vegetables or exotic fruit juices, fruit smoothies, and beverages rich in heat-sensitive but health-promoting micronutrients. However, not enough data are currently available to establish general processing criteria for PEF pasteurization of citrus juices and systematic research and evaluation of food safety and health aspects of PEF-processed foods directly after processing and during the acceptable shelf life is required on a case-by-case basis to assure food safety and obtain regulatory approval.