John D. Floros

Casting Antimicrobial Packaging Films and Measuring Their Physical Properties and Antimicrobial Activity

Antimicrobial film was extruded using LDPE resins and potassium sorbate powder; its tensile properties, transparency and antimicrobial activity were measured to examine the adaptability as a packaging material. The tensile properties were not affected significantly by the incorporation of potassium sorbate to 3% (w/w) in the film. However, transparency of the film decreased as the concentration of potassium sorbate increased. The antimicrobial film decreased the growth rate and maximum growth of yeast, and extended the lag period before mold growth was apparent. Therefore, it can prevent or reduce the rate of microbial spoilage in low viscosity liquids and on the contacted surface area of solid food products and thus may prolong the shelf life of food products when it is used as a packaging material.

Analysis and modeling of potassium sorbate diffusion through edible whey protein films

Abstract The mechanism of potassium sorbate release from edible whey protein films was investigated. Anomalous or non-Fickian diffusion mechanism with diffusional exponents ( n ) between 0.55 and 0.86 were found. A model describing the diffusion of potassium sorbate from whey protein films that swelled due to countercurrent diffusion of solvent was used to determine potassium sorbate and solvent diffusion coefficients. The diffusivities of potassium sorbate at 25°C ranged from 5.38 to 9.76×10 −11 m 2 /s, while those of the solvent from 1.29 to 7.43×10 −10 m 2 /s, depending on film composition.

Response surface modeling for the inactivation of Escherichia coli O157:H7 on green peppers (Capsicum annuum L.) by chlorine dioxide gas treatments.

The effects of chlorine dioxide (ClO2) gas concentration (0.1 to 0.5 mg/liter), relative humidity (RH) (55 to 95%), treatment time (7 to 135 min), and temperature (5 to 25 degrees C) on inactivation of Escherichia coli O157:H7 on green peppers were studied using response surface methods. A four-factor, central, composite, rotatable design was used. The microbial log reduction was measured as a response. A direct membrane-surface-plating method with tryptic soy agar and sorbitol MacConkey agar was used to resuscitate and enumerate ClO2-treated E. coli O157:H7 cells. The statistical analysis and the predictive model developed in this study suggest that ClO2 gas concentration, treatment time, RH, and temperature all significantly (P < 0.01) increased the inactivation of E. coli O157:H7. ClO2 gas concentration was the most important factor, whereas temperature was the least significant. The interaction between ClO2 gas concentration and RH indicated a synergistic effect. The predictive model was validated, and it could be used to determine effective ClO2 gas treatments to achieve a 5-log reduction of E. coli O157:H7 on green peppers.

Aseptic packaging machine pre-sterilisation and package sterilisation: statistical aspects of microbiological validation

Abstract Validation of the pre-sterilisation of an aseptic packaging machine and validation of the machine’s sterilisation of the packages require many evaluations, including a microbiological challenge. This paper describes the statistical considerations supporting the microbiological challenge, focusing especially upon the use of the most probable number (MPN) technique. The uniformity of the log count reduction (LCR) must be demonstrated for the standard sterilisation technique to be applied to the machine and to the packages. The sites in the machine expected to receive the least sterilisation effect must be identified, and all sites in the machine must be verified independently of each other. If the uniformity of sterilisation can be shown for the packages, these can be evaluated as large sets of replicates, achieving considerable precision of the LCR. The MPN technique requires evaluation of the improbability of the results and must use credible estimates of the confidence interval. These techniques are thoroughly discussed, and a spreadsheet routine is demonstrated that incorporates the techniques.

Determining the diffusivity of sodium hydroxide through tomato and capsicum skins

Abstract An experimental method to measure the effective diffusivity of chemical substances through the skin (exocarp) of fruits and vegetables is described. A circular disk of skin was placed in a specially constructed diffusion cell to separate two flowing solutions. Changes in conxentration with time were monitored and the diffusion properties were estimated from Ficks diffusion equation. The calculations were based on results taken either in the early transient period or in the late period of steady-state flow. The method and the design of the diffusion cell allowed measurements with negligible edge effects and rapid detection of errors due to leaks or contamination. Diffusion characteristics were determined for the penetration and movement of sodium hydroxide through the skins of tomatoes and pimiento peppers. The diffusivity through tomato skin was 2·0 × 10−8cm2s−1 and through pimiento pepper skin 5·5 × 10−8cm2s−1 at 72°C. Results were reproducible and agreed with previous studies.

COMPUTER-AIDED MICROBIAL IDENTIFICATION USING DECISION TREES

Abstract This paper will demonstrate the utility of using machine learning methods to develop more efficient microbial identification (ID) techniques. The use of computer algorithms to create new decision trees can improve efficiency and increase systematization in the field of microbiology. Preliminary results indicate that decision tree algorithms can create new structures that require fewer tests on average to reach a positive identification of an unknown organism. Including test time and cost factors can make further improvements, resulting in systems that are more time-efficient and/or cost-effective. Machine learning techniques can also create customized ID systems for specific applications. This paper will explain the induction of decision trees and show examples of their use in microbial ID.

Genetic algorithms and fuzzy theory for optimization and control of food processes

Abstract Genetic Algorithms are search algorithms seeking improved (optimized) performance by mimicking the mechanics of natural selection and genetics. Fuzzy set theory imitates the human way of judging by using linguistic descriptions and criteria. A combination of the two methods may assist the optimization and on-line control of food processes. The structure of Genetic Algorithms and the way optimization is implemented through them are presented. Emphasis is placed on classifier systems. Furthermore, fuzzy set theory and its applications in fuzzy control are discussed. The important aspects of combining the two methods (Genetic Algorithms and Fuzzy Control) for simultaneous optimization and control of food processes are investigated. Some potential applications in the food industry are briefly discussed.

Multiresponse optimization by a normalized function approach

Abstract A simple new method, called Normalized Function Approach (NFA), was developed for simultaneously optimizing systems defined by several output variables (response functions) and a common set of input variables (factors). For each original response function the difference between the estimated response and its individual optimum was evaluated and normalized over the experimental space. The individually normalized functions were then weighted for their importance and combined into an overall function using a “sum of squares” approach. Simultaneous optimization was completed by minimizing the overall function. Some numerical examples were considered to introduce the new method and several reported applications in the food area were used for verification and comparison with other multiresponse optimization methods.

Back propagation neural networks: Theory and applications for food science and technology

Abstract Neural networks (NNs), a class of powerful information processing techniques, may be used to characterize complex nonlinear relationships such as those found in food processing operations. A brief overview of the various NNs is presented. In depth discussion of back propagation networks (BPNs), a simple technique with wide range of applicability, follows. The mechanics of constructing and using BPNs including their characteristics, definition, training and testing are covered. The limitations of BPNs and practical guidelines for their implementation are also discussed. Current and potential applications of BPNs in food science and technology are presented with examples in process modeling, optimization and control, food and sensory analysis, and product formulation.

Problems Associated With the Processing of Cucumber Pickles: Softening, Bloater Formation and Environmental Pollution

ABSTRACT The traditional process of cucumber fermentation and storage was critically reviewed. Enzymatic and non-enzymatic softening, bloater formation and environmental pollution due to high salt brine disposal were identified as the most important problems. The enzymes and acids involved in softening, and the mechanism of bloater formation were examined and some solutions (firming agents, cucumber washing, brine purging, homofermentative starters and controlled fermentation) were presented. Methods to reduce environmental pollution by salt recovery, brine recycling, closed tank controlled/anaerobic fermentation and low-salt natural fermentation and storage were discussed and debated.