A juice plant in Brazil wants to replace the flash pasteuriser on its not-from-concentrate orange juice line with high pressure processing (HPP), and to treat its clear sugar syrup in a UV-C reactor. Both suppliers promise a 5-log reduction of pathogens. The supermarket buying the juice asks for something else: the validation file. A brochure figure, measured on another product in another laboratory, is not evidence for this juice, this syrup or this plant. Novel food process validation produces that evidence, and the logic is the same for HPP, pulsed electric fields (PEF) and UV-C.
In short
- Novel food process validation shows that a non-thermal process, operated at its critical limits, achieves the required log reduction of the target pathogen in the actual product.
- Design for the worst case: the most protective formulation, the weakest permitted settings and the most resistant strains, usually as a multi-strain cocktail.
- A surrogate such as Listeria innocua, a non-pathogenic Escherichia coli strain or Enterococcus faecium is acceptable only once shown to be at least as resistant as the pathogen under that process, in that food.
- Non-thermal survivor curves often tail, so measure the target log reduction directly instead of extrapolating from early data.
- HPP at 400 to 600 MPa and chilled temperature does not inactivate bacterial spores; low-acid HPP foods still need refrigeration and other hurdles.
What does novel food process validation involve?
Novel food process validation is the gathering of scientific evidence that a process such as HPP, PEF or UV-C, operated within defined limits, will consistently achieve a stated level of control of a named hazard in a named product. The Codex Guidelines for the Validation of Food Safety Control Measures (CXG 69-2008) apply exactly as they do to a retort; only the physics changes.
A log reduction is the fall in microbial numbers on a base-10 logarithmic scale: a 5-log reduction takes 100,000 cells to 1. The performance criterion, the reduction the process must deliver, comes from the hazard analysis, the expected contamination and any legal requirement. Juice processes commonly use a 5-log reduction of the most resistant pertinent pathogen, and some national rules, for example in the United States, make that mandatory. Check your national legislation. A complete validation file normally contains:
- the target pathogens and the performance criterion;
- the critical parameters and their proposed limits;
- laboratory challenge studies with the pathogens at worst-case conditions;
- confirmation on production equipment, usually with surrogates or dose indicators;
- shelf-life studies showing that survivors, including injured cells, do not recover and grow, also under mild temperature abuse;
- monitoring and verification procedures tied to the validated limits.
How do HPP, PEF and UV-C inactivate microorganisms?
All three damage vegetative cells near ambient temperature, but each delivers its effect differently, so each has its own critical parameters and blind spots.
| Process | Mode of action | Typical conditions | Critical parameters | Main limitation |
|---|---|---|---|---|
| HPP | Isostatic pressure damages membranes and proteins | 400 to 600 MPa for about 1 to 6 min | Pressure, hold time, temperature, water activity | Spores survive; low water activity protects cells |
| PEF | Electroporation of cell membranes by microsecond pulses | About 10 to 40 kV/cm | Field strength, specific energy, outlet temperature, flow, conductivity | Spores and many enzymes resist; arcing |
| UV-C | Photons near 254 nm form pyrimidine dimers in DNA | Fluence in mJ/cmΒ² | Delivered fluence, UV transmittance, flow, lamp output | Poor penetration in absorbing liquids; shading |
HPP. Pressure is transmitted uniformly and almost instantly through the pack, so there is no pressure cold spot. Bacterial spores, however, are not inactivated at chilled or ambient temperature, and high sugar or salt protects vegetative cells. Low-acid HPP products, such as avocado dips, keep their spores and need strict refrigeration and a shelf life set with non-proteolytic C. botulinum in mind, which can grow from about 3 Β°C.
PEF. Electroporation is the formation of pores in cell membranes when an electric field drives the membrane voltage past a critical value. Treatment intensity is best described by specific energy, the electrical energy delivered per kilogram of product, where Ο is conductivity, E field strength, Ο pulse width, n the number of pulses and Ο density:
W = Ο Γ EΒ² Γ Ο Γ n / Ο
Orange juice with Ο = 0.3 S/m given 20 pulses of 2 Β΅s at 30 kV/cm (3 Γ 10βΆ V/m) receives 0.3 Γ (3 Γ 10βΆ)Β² Γ 2 Γ 10β»βΆ Γ 20 / 1000 β 108,000 J/kg, or 108 kJ/kg (density about 1000 kg/mΒ³). With a heat capacity of about 3.9 kJ/kgΒ·K, that is a temperature rise of about 28 Β°C. Microbicidal PEF is not free of heat.
UV-C. Low-pressure mercury lamps emit mainly at 253.7 nm, close to the absorption peak of DNA. Fluence, the UV dose, is irradiance multiplied by time: 2 mW/cmΒ² for 10 s gives 20 mJ/cmΒ². Absorbance is the problem: intensity falls exponentially with depth (the Beer-Lambert law), so in a clarified juice with an illustrative absorption coefficient of 15 per cm (base 10), intensity 1 mm from the lamp is 10^(β1.5), about 3 % of the surface value. Milk and cloudy juices absorb even more strongly.
What are worst-case conditions in a validation study?
Worst-case conditions are the combination of product and process settings, within specification, that gives the least lethality. Validating there means every batch made inside the limits receives at least the demonstrated reduction.
- HPP: lowest pressure and shortest hold at the control limits, lowest water activity (highest soluble solids), highest pH, and whichever end of the permitted temperature range trials show to be less lethal.
- PEF: lowest field strength and specific energy, lowest inlet temperature, and the highest conductivity and particle content allowed, which reduce uniformity and can limit the field the generator sustains.
- UV-C: highest product absorbance and turbidity, highest flow rate (shortest exposure), and lamp output at its alarm or end-of-life value with fouled sleeves.
- All processes: the most protective recipe and ingredient lot permitted, such as maximum pulp or fat.
Timing of enumeration matters too. In acidic juices, pressure-injured cells often die during chilled storage. That die-off should be credited to the process only if it is separately demonstrated and controlled; otherwise, count survivors immediately after treatment.
How do you choose challenge organisms and surrogates?
Use the pathogens of concern in the laboratory, as a cocktail of several strains including outbreak isolates and strains known to resist the process, and use a surrogate where pathogens cannot be handled, typically on production equipment. A surrogate is a non-pathogenic organism whose resistance to the specific process, in the specific food, is equal to or greater than that of the target pathogen.
| Target | Surrogate examples | Notes |
|---|---|---|
| Listeria monocytogenes | Listeria innocua | Widely used; confirm resistance in the product |
| E. coli O157:H7 | Non-pathogenic E. coli strains | Used in juice HPP, PEF and UV work; strains vary |
| Salmonella in low-moisture foods | Enterococcus faecium (NRRL B-2354) | Established for thermal pasteurisation of nuts and dry foods |
| Proteolytic C. botulinum | Clostridium sporogenes | Thermal processing; not automatically valid for pressure with heat |
| UV reactor fluence | MS2 bacteriophage, Bacillus subtilis spores | Biodosimeters in water treatment |
Resistance to heat does not predict resistance to pressure, fields or UV, so a surrogate proven for one technology or matrix is not automatically valid for another. Grow cells in a standardised way, usually to stationary phase, which is generally more resistant than exponential growth, and acid-adapt them for acidic products.
Inoculate high enough to measure the target: to show 5 logs with a detection limit of about 1 log CFU/g, start at about 6.5 to 7 log CFU/g or more. CFU, colony-forming units, counts cells able to form colonies on agar. Sublethal injury is damage that stops cells growing on selective media straight after treatment but allows later recovery, so use a non-selective resuscitation step before selective counting. Ignoring injury overstates the log reduction.
How should inactivation kinetics be modelled?
Model the data you measured, and never extrapolate a log reduction beyond them. Classical D values assume a straight line on a log survivor plot, but non-thermal processes often show shoulders, a lag before kill begins, or tails, a resistant or protected fraction. The Weibull model handles both:
logββ(N/Nβ) = β(t/Ξ΄)α΅ and tβ = Ξ΄ Γ n^(1/p)
Here Ξ΄ is the time to the first decimal reduction and p the shape parameter: p = 1 is log-linear, p below 1 means tailing and p above 1 means a shoulder; tβ is the time for n decimal reductions.
Worked example
An HPP trial on a cold-pressed juice at 550 MPa gives mean reductions of 2.4 logs at 1 min, 3.4 logs at 2 min and 4.8 logs at 4 min.
Fit the 1 and 4 min points: p = log(4.8/2.4) / log(4/1) = 0.301 / 0.602 = 0.50. Then 2.4 = (1/Ξ΄)^0.5, so Ξ΄ = 1/2.4Β² β 0.174 min. Check at 2 min: (2/0.174)^0.5 β 3.4 logs, matching the data.
Time for 5 logs = 0.174 Γ 5^(1/0.5) = 0.174 Γ 25 β 4.3 min. A straight line from the first point (2.4 logs per minute) would predict 5 logs in about 2.1 min, half the real requirement. Because 4.3 min lies beyond the data, even the Weibull figure must be confirmed by a direct test at the proposed hold time.
Replicates matter as much as the mean. If a direct test at 4.5 min gives a mean of 5.1 logs with replicates from 4.6 to 5.6, the mean passes but the process does not consistently deliver 5 logs. Judge it on the lower confidence limit, not the mean.
For PEF, or any process that heats the product, run a thermal-only control with the same temperature history so that heat is not mistaken for a field or pressure effect. Inactivation kinetics and sublethal injury are covered in depth in Advanced Food Science.
How does a validated process become a controlled one?
Each production critical limit should trace back to the worst-case conditions under which the target reduction was demonstrated, and each should be monitored for every batch or cycle.
- HPP: pressure and hold time logged per cycle with calibrated transmitters, process water temperature where specified, pH and soluble solids per batch, and physical segregation of treated packs.
- PEF: voltage, current and pulse shape logged in real time, specific energy, inlet and outlet temperatures, flow and product conductivity, with diversion when any falls outside limits.
- UV-C: UV sensor readings against the validated minimum, flow rate, product transmittance per batch, lamp hours and sleeve cleaning.
Verification covers calibration, periodic microbiological tests and record audits. Revalidate when the formulation, pack, equipment, supplier or shelf life changes. Regulatory status also varies: some jurisdictions require pre-market approval for certain novel processes, and terms such as “fresh” or “raw” are regulated differently, so check your national legislation. The Advanced Food Technology course includes a full validation plan for a high pressure processed juice.
Frequently asked questions
Does HPP kill bacterial spores?
Not at the chilled or ambient temperatures used commercially. HPP at 400 to 600 MPa inactivates vegetative bacteria, yeasts and moulds, but bacterial spores survive. Combining pressure with high temperature can inactivate spores, but that approach has very limited commercial use. Low-acid HPP foods therefore need refrigeration and a shelf life set with non-proteolytic Clostridium botulinum in mind.
Why is UV-C hard to use on juices and milk?
UV-C must reach the microbes, and many liquids absorb it strongly at 254 nm. In cloudy juices, milk or liquid egg, most UV is absorbed within a fraction of a millimetre of the lamp, so deeper microbes receive little dose. Reactors therefore use thin films or turbulent and vortex flow, and validation measures the delivered dose biologically.
Can a surrogate replace the pathogen in a validation study?
Only once it has been shown to be at least as resistant as the target pathogen under the same process and in the same food. Surrogates such as Listeria innocua or non-pathogenic E. coli strains are mainly used on production equipment, where pathogens must not be introduced. The core laboratory evidence normally comes from a multi-strain cocktail of the pathogens themselves.
What log reduction does a novel process need?
It depends on the hazard analysis, the expected contamination level and any legal performance standard. Juice processes commonly target a 5-log reduction of the most resistant pertinent pathogen, and some countries make that mandatory. Whatever the figure, demonstrate it at worst-case conditions with enough inoculum to measure it, rather than extrapolating.
Is PEF really a non-thermal process?
Partly. PEF kills cells mainly by electroporation, but microbicidal treatments typically deliver tens of kilojoules per kilogram or more, and 100 kJ/kg raises juice temperature by about 25 Β°C. Commercial systems often use that heat deliberately, then cool quickly. Validation must therefore record inlet and outlet temperatures and include a thermal-only control, so that the lethality credited to the electric field is real.
Next step. The Advanced Food Technology course covers the physics of HPP, PEF, UV-C, cold plasma and ultrasound, Weibull and other non-log-linear kinetics, surrogate selection and worst-case study design. It finishes with a proctored final assessment and an ASC certificate, and you can see all eleven food science and technology courses.
Sources. Codex Alimentarius Commission, Guidelines for the Validation of Food Safety Control Measures (CXG 69-2008), available from the Codex Alimentarius website. EFSA Panel on Biological Hazards, The efficacy and safety of high-pressure processing of food, EFSA Journal (2022), from EFSA. Institute of Food Technologists for the US FDA, Kinetics of Microbial Inactivation for Alternative Food Processing Technologies (2000). National Advisory Committee on Microbiological Criteria for Foods, Parameters for Determining Inoculated Pack/Challenge Study Protocols, Journal of Food Protection (2010). P. J. Fellows, Food Processing Technology: Principles and Practice, 5th edn (Woodhead Publishing, 2022).
This article is general guidance and is not a substitute for the applicable standard, legislation or the advice of a qualified food safety professional.