Journal of Food Engineering

Measuring food volume (portion size) is a critical component in both clinical and research dietary studies. With the wide availability of cell phones and other camera-ready mobile devices, food pictures can be taken, stored or transmitted easily to form an image based dietary record. Although this record enables a more accurate dietary recall, a digital image of food usually cannot be used to estimate portion size directly due to the lack of information about the scale and orientation of the food within the image. The objective of this study is to investigate two novel approaches to provide the missing information, enabling food volume estimation from a single image. Both approaches are based on an elliptical reference pattern, such as the image of a circular pattern (e.g., circular plate) or a projected elliptical spotlight. Using this reference pattern and image processing techniques, the location and orientation of food objects and their volumes are calculated. Experiments were performed to validate our methods using a variety of objects, including regularly shaped objects and food samples.

Dielectric properties of 16 process cheeses were determined over the frequency range 0.3–3 GHz. The effect of temperature on the dielectric properties of process cheeses were investigated at temperature intervals of 10 °C between 5 and 85 °C. Results showed that the dielectric constant (ε′) decreased gradually as frequency increased, for all cheeses. The dielectric loss factor (ε″) decreased from above 125 to below 12 as frequency increased. ε′ was highest at 5 °C and generally decreased up to a temperature between 55 and 75 °C. ε″ generally increased with increasing temperature for high and medium moisture/fat ratio cheeses. ε″ decreased with temperature between 5 and 55 °C and then increased, for low moisture/fat ratio cheese. Partial least square regression models indicated that ε′ and ε″ could be used as a quality control screening application to measure moisture content and inorganic salt content of process cheese, respectively.

There is a large market for microwave foods, with one of the most prevalent growth areas being low-density bakery products. The dielectric constant and loss factor of Madeira cake batter and its constituents (sugar, cake concentrate, margarine, flour, egg) were measured at a range of moisture contents (0.429–1.000 kg kg−1, dry basis), temperatures (20–80 °C), and over a frequency range of 915–2450 MHz. In general, for a given material, the dielectric properties of samples increased with increasing frequency. Irrespective of frequency and sample moisture content, the dielectric constant of batter and flour was relatively independent of sample temperature, whereas that of the sugar samples increased with increasing temperature, particularly within the lower moisture range. At 2450 MHz the loss factor of batter and flour samples decreased by a limited amount as temperature was increased, whereas at 915 MHz they were relatively unaffected. The sugar samples showed a significant decrease in loss factor with increasing temperature, irrespective of frequency. During microwave baking, the dielectric properties of Madeira cake batter initially increased sharply and then decreased steadily until the end of the baking process.

Our formerly published experimental results of studies on inactivation of Bacillus subtilis As 1.173 spores by high hydrostatic pressure and heat using design of experiments showed that the optimum process parameters for a six log-cycle reduction of spores of B. subtilis were obtained as: pressure, 576.0 MPa; temperature, 87 °C; and pressure holding time, 13 min. Based on the results, response surface methodology (RSM) was performed in the present investigation, to determine effects of food composition (soybean protein, bean oil, sucrose) and pH on the inactivation of B. subtilis As 1.173 spores by high pressure and mild heat were studied. A quadratic predictive model was built for the effects of food compositions and pH on the reduction levels of B. subtilis spores by HPP using RSM. The experimental results showed that the reduction of B. subtilis spores in buffer and foods differed depending on the HPP treatment process parameters. The soybean protein (P = 0.0005), sucrose (P = 0.0003), and pH (P = 0.0080) significantly influenced the reduction of B. subtilis spores, and the effect of bean oil on the reduction of B. subtilis spores was only slightly significant (P = 0.0810). Moreover, the adequacy of the model equation for predicting the reduction of B. subtilis spores was verified effectively using experimental test data that was not used in the development of the model.

New Zealand chestnuts grow relatively large by comparison with Asian or American varieties and have the potential to be a lucrative crop. However they are susceptible to fungal rots and need to be dried at relatively low temperatures in order to prevent significant spoilage. Drying curves for New Zealand chestnut variety ‘1015’ were obtained at 30 °C, which was the maximum drying temperature that did not cause significant quality degradation. Two distinct falling-rate drying periods were observed: an initial period during which a relatively high drying rate decreased rapidly, followed by a period of lower drying rates which decreased only gradually. The first period most likely corresponded to the time required for the surface region to reach moisture equilibrium, since it was clearly affected by external factors such as air movement and the absence or presence of moisture barrier(s). The second period represented the diffusion-controlled process, and was largely unaffected by external factors. The mean apparent diffusivity of whole chestnuts at 30 °C was 5.1 × 10−11 m2 s−1.

Orange waste resulting from juice processing that contains a high level of fermentescible sugars was utilized as a basal medium at 2% (w/v) for the production of α-amylase (E.C.3.2.1.1) using a fungal strain of Aspergillus oryzae (Ahlburg) Cohen 1042.72, at 30 °C on a rotary shaker. The production of the α-amylase, proteins and biomass was achieved using the statistical method of [Plackett–Burman design at N=12, namely 12 experiments and 11 factors (7 real values and 4 dummy)]. Statistical analysis showed the following results: (a) the variation of starch concentration (0 at 2%), of yeast extract (0 at 0.5%) and of “Corn steep liquor” (0 at 2%) has a positive effect which is highly significant on the production of α-amylase and proteins with the following probabilities respectively (p⩾80–95%, p⩾80–95%, p⩾70–99%). (b) as for biomass, it seems that it is positively affected by the variation of starch (p⩾95%), “Corn steep liquor” (p⩾95%) and agitation (150–200 rpm) (p⩾80%). (c) the salt utilized (MgSO4, 7H2O 0 at 0.5 g/l, MnSO4 0 at 0.1 g/l, FeSO4 0 at 0.1 g/l, ZnSO4 0 at 0.02 g/l) showed a negative effect which is highly significant (p⩾80%) on the production of proteins and without effect on the two previous responses (enzyme production and biomass). On conclusion, the statistical analysis allowed the selection of factors such as starch, yeast extract, “Corn steep liquor” and a low concentration of salt.

Electron beam irradiation can inhibit pathogens in fresh produce while maintaining its quality. However, proper dose determination in complex-shaped produce is difficult, and needs further evaluation to ensure adequate irradiation treatment. The objective of this study was to establish the best treatment for irradiation of a broccoli head by comparing simulated dose distributions from Monte Carlo simulation and computed tomography (CT) scan data with those measured using standard dosimeters (alanine pellets and radiochromic films). Values of 3-D geometry and component densities of a broccoli head were entered into a radiation transport code (MCNP5) to simulate dose distribution. The broccoli head was irradiated with a 10 MeV double beam (top and bottom) linear accelerator. Dose levels ranged from 1 to 3 kGy.Several irradiation strategies were simulated (position of the product related to the beam entrance and, beam sources) and compared with the experimental data. The best results were obtained when the broccoli was irradiated at a 132.5° angle using the double beam configuration. The Dmax/Dmin, a measure of the dose uniformity, was 1.72 around the floret and, 2.22 for the whole broccoli.These results provide valuable information for irradiation treatment planning of complex-shaped fresh produce, thus ensuring that the product is uniformly exposed to the target dose while preserving its quality attributes.

The performance of an extrusion process designed to convert acid casein to sodium caseinate was evaluated relative to screw speed, temperature and feed rate; their effects on mixing, conveying, specific energy consumption (SEC), pressure, product temperature and end product physicochemical characteristics were considered. The effect of the process variables were assessed in terms of resulting residence time distribution (RTD) and associated parameters, and compared to theoretical models in an attempt to describe the flow behavior in the extruder. The process variables did not have a significant effect on physicochemical properties, however, they strongly affected RTD patterns, SEC, pressure generated and product temperature. A detailed analysis of the process is presented, indicating that the twin-screw extruder reactor is well described by a plug flow/perfect mixing model. The process produced a well mixed homogeneous product of comparable quality to conventionally manufactured caseinate and the performance analysis carried out provides a solid basis for the further optimization of such a process.

Achu is a paste traditionally made from taro corms by boiling, peeling and pounding in a mortar to a smooth and homogeneous consistency. The present study aims to define alternative and rapid methods to evaluate the textural hardness of achu reconstituted from taro flour. The creep recovery behaviour of achu samples made from flour of five varieties of taro was evaluated on a controlled-stress rheometer. The functional properties of the taro flours, the sensory analysis of hardness and the acceptability of reconstituted achu were also evaluated. The results showed significant variation in the functional properties of the flours. The rheological properties of the achu, the sensory hardness and overall acceptability were also significantly influenced by the variety. In general, flour which absorbed more water tended to produce achu with high level of newtonian compliance and retarded elastic compliance. The creep steady state shear compliance and the viscoelasticity index of achu, were significantly (P < 0.05) correlated to the sensory hardness.

Several nonlinear models (Gompertz, Logistic-1, Logistic-2, Richards, Stannard and Schnute models) were compared to describe Yersinia enterocolitica and aerobic bacteria inactivation in Turkish Feta cheese during the storage period. They were compared statistically by using the F-test. The Gompertz model was decided as a sufficient model, which describes the inactivation of both Y. enterocolitica and aerobic bacteria. The biological parameters were calculated from the modified Gompertz model. The effect of starter culture (Streptococcus cremoris), salt (150 or 200 g/l) and initial inoculum level of Y. enterocolitica on the phase of disappearance (λ) and the rate of reduction (μ) of Y. enterocolitica and aerobic bacteria were determined. Addition of starter culture had a significant (P<0.05) inhibitory effect on both Y. enterocolitica and aerobic bacteria in Turkish Feta cheese. However, salt and inoculation rate of Y. enterocolitica had no significant (P>0.10) effect on both λ and μ of microorganisms.

Aim of this work was the optimization of xylitol production by Debaryomyces hansenii UFV-170, which already proved to be a new promising xylitol-producing yeast. Two sets of batch bioconversion tests were carried out on synthetic medium, according to two joined 33 and 32 type full factorial designs, selecting the initial xylose concentration (S0), rotational speed (v) and starting biomass concentration (X0) as independent variables and the maximum xylitol concentration (P), xylitol yield on consumed xylose (YP/S), volumetric productivity (QP) and specific productivity (qP) as response variables. The collected results were then worked out by response surface methodology (RSM). Overall optimization, conducted by overlaying the curves of the responses under investigation, allowed us to point out an optimal range of the independent variables within which the four responses were simultaneously optimized. The point chosen as representative of this optimal area corresponded to S0 = 156 g L−1, v = 280 rpm and X0 = 6.4 g L−1, conditions under which the model predicted P = 116.25 g L−1, YP/S = 0.77 g g−1, QP = 1.49 g L−1 h−1 and qP = 0.16 g g−1 h−1.

This study was carried out to determine the effect of moisture content on some physical properties and mechanical behaviour under compression load of faba bean grains. Four levels of moisture content ranging from 9.89% to 25.08% d.b. (dry basis) was used. The average length, width, thickness, geometric mean diameter, unit mass of grain, sphericity, thousand grain mass and angle of repose ranged from 18.40 to 19.77 mm, 12.54 to 13.66 mm, 7.00 to 8.03 mm, 11.68 to 13.01 mm, 1.147 to 1.301 g, 63.47% to 65.78%, 1140.15 to 1332.67 g and 13.94° to 18.58° as the moisture content increased from 9.89% to 25.08% d.b., respectively. As the moisture content increased from 9.89% to 25.08% d.b., the bulk density was found to decrease from 419.59 to 381.60 kg/m3, whereas the grain volume, true density, porosity and surface area were found to increase from 0.998 to 1.099 cm3, 1151.33 to 1206.21 kg/m3, 63.09 to 67.21% and 4.29 to 5.31 cm2, respectively. The static and dynamic coefficients of friction on various surfaces, namely, galvanized metal, chipboard, mild steel, plywood and rubber also increased linearly with an increase in moisture content. The mechanic properties of faba bean were determined in terms of average rupture force, specific deformation and rupture energy along X-, Y-, and Z-axes. Specific deformation and rupture energy of the faba bean grains generally increased in magnitude with an increase in moisture content, while rupture force decreased for compression along X-, Y-, and Z-axes. The highest rupture force, specific deformation and rupture energy in all moisture content levels were obtained for faba bean grains loaded along the Z-axis.

Kinetics of osmotic dehydration (OD) and effects of sucrose impregnation on thermal air-drying of pumpkin slices were investigated. A simplified model based on the solution of Fick’s Law was used to estimate effective diffusion coefficients during OD and air-drying. In order to take into account shrinkage, average and variable thicknesses were considered. Pumpkin slices were dehydrated in sucrose solutions (40%, 50% and 60%, w/w, 27 °C). The effective water diffusion coefficients were higher than the sucrose, and low diffusivity dependence with solution concentration was observed. Samples non-treated and pre-treated in 60% osmotic solutions during one hour were dried in a hot-air-dryer at 50 and 70 °C (2 m/s) until equilibrium was achieved. Pre-treatment enhanced mass transfer during air-drying. Great volume reduction was observed in pre and non-treated dried samples. Using variable thickness in the model diminished the relative deviations between predicted and experimental OD and drying data.

Inactivation kinetics (D and z values) of Salmonella typhimurium under high pressure carbon dioxide (HPCD) from 1.51 to 7.56 MPa at 35°C was investigated in model solutions of brain heart infusion broth (BHIB) and physiological saline (PS). The come-up time to reach a pressure exerts an important effect on the survival of S. typhimurium. The inactivation rates increased with pressure, suspending medium and exposure time. Inactivation followed first order reaction kinetics, with specific inactivation rates (k) and decimal reduction times (D) that varied from 0.0975 to and from 23.63 to 1.48 min, respectively. The pressure dependence of the S. typhimurium specific inactivation rates can be described by the zp value depending on the model solutions BHIB and PS.

One-dimensional 23Na NMR imaging (MRI) has been used to follow the ingress of sodium ions into post-rigour porcine muscle during brining. The ingress is well described by Fick's second law, with an inter-diffusion coefficient that decreases exponentially with the sodium ion content (between 8 and 20 mg g−1 cured pork). It is suggested that this decrease is enhanced by the increasing electrostatic charge of the meat proteins and by the dissolution of myosin into solution that accompanies the increase of the NaCl concentration in cured pork.

Cryomechanical freezing is recommended for delicate products, with poor mechanical resistance (shrimps, raspberries, strawberries) or those that change their appearance during freezing (chicken, mushrooms, shrimps). The aim of this work is to study the quality aspects related to the application of cryomechanical freezing compared to the use of a conventional mechanical freezer. In order to determine whether or not the use of the combined freezer provokes an improvement in the quality of the final product, different kinds of foodstuffs were chosen and analysed before and after freezing by both methods. The selected products were: chicken escallops, hamburgers, strawberries, asparagus and mushrooms. The quality parameters analysed for this purpose were: drip loss during thawing, texture and colour. The variation in mechanical resistance was also evaluated as a function of the immersion time to determine the hardness of the protective crust formed by the liquid N2 pretreatment.

The densities and viscosities of ternary NaCl–glucose syrup–water systems were measured in the 283.1–298.1 K temperature range and the water activities were measured at 298.1 K. NaCl and glucose syrup molalities ranged from 0 to 6.0 mol kg−1 and from 0 to 1.91 mol kg−1, respectively. “Dextrose equivalent” (DE) glucose syrup value was 21 and the median molecular weight was 1.007 (kg mol−1). Experimental density, viscosity and water activity data were fitted by a five-parameter correlation, an eight-parameter polynomial and a five-parameter polynomial correlation, respectively with deviations of less than 0.5%, 2.3% and 0.9%, respectively. The presence of NaCl was found to be responsible for excess volume. There was a major contribution of interaction terms in the calculation of the ternary viscosity, particularly for high viscosity. While these ternary systems are used in the dehydration–impregnation soaking (DIS) process, the technological consequences of the findings were discussed in terms of DIS process optimization, mass transfers, solution management.

This work describes an innovative process for the extraction of lycopene from mated cultures of Blakeslea trispora using supercritical carbon dioxide extraction (SCE) containing acetone as an entrainer. The results obtained are compared to conventional solvent extraction. Incorporation of entrainer improved the recovery of lycopene from 73% to 85% by SCE. Box–Behnken design was used to optimize the operating conditions for maximum extraction of lycopene from the biomass. The parameters optimized were temperature (45–65 °C), pressure (300–400 bar), and extraction time (1.0–2.0 h) at a fixed flow rate of 1 ml/min of carbon dioxide and entrainer at 1 ml/g biomass. The optimum conditions were found to be 1.1 h extraction at 52 °C and 349 bar, resulting in a 92% recovery of lycopene. The use of supercritical fluid extraction along with an entrainer is therefore recommended as a better option for enhancing the recovery of lycopene from B. trispora.

Finite element modeling was used in combination with 2D MRI temperature mapping to calculate fluid to particle convective heat transfer coefficients (hfp) across the surfaces of a carrot particle being heated with 80 °C water with an average velocity of 4.4 cm/s. Heat transfer in the region of interest (image acquired from center of sample) was essentially two-dimensional because of the length of the carrot sliver. The hfp at each surface was determined by trial and error matching of temperature contours in the model to those in the MRI image. Calculated values of hfp (117–389 W/m2 K) were within the range of those reported in the literature. The major advantages of this method are that the actual shape of the food particle is used in the model, and it can be utilized in applications where the particle does not heat uniformly from all directions.

A 2D fouling model for milk pasteurizer using plate heat exchangers (PHEs) was implemented and validated. The model emphasis was placed on fouling prediction based on the hydrodynamic and thermodynamic performances of PHEs in 2D environment. Different types of multichannel PHEs, i.e. 12-channel PHE 1 and 20-channel PHE 2 with both counter-current and co-current flows were applied to validate the model by visualization and quantification of milk deposit developed in the fluid channels. The predicted mass deposit values at each channel were in good agreement with the experimental data with the average prediction error of 0.4 g or 10.9% (R2 = 0.98) for PHE 1. The theoretical values of temperature and mass deposit for a plate of PHE 2 were closer to the actual measurements with errors of 1.6 °C or 2.1% and 0.06 g or 6%, respectively. The aggregation rate of unfolded protein was observed to increase exponentially with increasing wall temperature, due to a reduction in flow rate and heat transfer coefficient. The predicted milk deposit distribution on the plate surface is expected to determine the sensor location appropriate for on-line monitoring and furthermore, optimization of the thermal process for minimizing milk fouling.

Mechanical evolution of tofu curd in gelation was investigated using low-power ultrasound and textural analysis. Two independent ultrasonic parameters, velocity and attenuation, were measured at the frequency 1 MHz as a function of time after addition of the calcium sulphate (CaSO4·2H2O) coagulant to heated soya milk. The responsive ultrasonic velocity has a plateau in the beginning of gelation and tends to a lower steady state after the formation of tofu gels. Ultrasonic attenuation exhibits first-order kinetics that matches the development of firmness revealed by textural analysis. Low-power ultrasound explores the formation of tofu gels in the aspects of pre-gelation processes, protein aggregation in gelation, and mechanical evolution in the gel at the post-gelation stage.

The activity and kinetics of carrot peroxidase were determined by using pyrogallol, guaiacol and o-dianisidine as hydrogen donors and peroxidase inactivation was studied by thermal and microwave heating. The kinetics of peroxidase showed characteristics which were dependent upon the identity and concentration of the hydrogen donor used. With pyrogallol and guaiacol, the true Km was found to be 0.34 and 1.4 mM for hydrogen peroxide, respectively, whereas the apparent Km with o-dianisidine was 7.7 × 10−3 mM. The lowest Km and highest Vmax/Km with o-dianisidine exhibited the greater tendency of the enzyme toward hydrogen peroxide and the specificity of the competing substrate, o-dianisidine. Thermal treatment of carrot peroxidase was done in the range of 35–75 °C for 0.5–180 min. Inactivation kinetics of peroxidase showed a biphasic first-order model, while at 75 °C, peroxidase showed monophasic first-order behaviour. Kinetic parameters, k and Ea, were determined for heat labile and heat resistant fractions of peroxidase. Biphasic behaviour of enzyme inactivation was observed for the microwave treatment at 70 and 210 W, whereas at 350 and 700 W microwave powers enzyme inactivation was monophasic. Microwave heating was found to be more effective for inactivating the enzyme than thermal treatment and additionally vitamin C retention was higher in microwave treated samples compared to heat treatment.

The rheological properties of corn starch–milk–sugar (CMS) paste were investigated using a concentric cylinder viscometer. Three types of sugars, namely, glucose, fructose and sucrose were used to prepare the paste. The paste was prepared at three different starch concentrations, 2, 4 and 6 wt.%, while the sugar concentrations of the three types were varied up to 16 wt.%. Milk with three different fat contents was used in preparation of the CMS paste. At constant starch and sugar concentrations and using skim milk, glucose was found to have the largest effect on the apparent viscosity of the CMS past. On the other hand, the fructose has the lowest effect on the CMS paste viscosity. It was also demonstrated that the pastes prepared with full cream milk have higher viscosity than the ones prepared using skim milk. The rheological CMS paste properties were modelled using Herschel–Bulkley model. The model was adequate and accurate.

High hydrostatic pressure kinetics (D, z and k values) of aerobic bacteria (AB) and Listeria monocytogenes inactivation were determined in brain heart infusion broth (BHIB) and foods. The D and k values of L. monocytogenes ranged from 12.1 to 1.14 min and from 0.1907 to 2.0202 min−1, respectively in BHIB at a pressure range 200–700 MPa. D values were about 3.04 and 2.43 min for AB and L. monocytogenes, respectively, in milk at 600 MPa, 2.13 and 1.52 min, respectively, in peach juice, and 1.24 and 0.87 min, respectively, in orange juice. The k values for L. monocytogenes in orange juice (ranged from 0.8024 to 2.6471 min−1) were greater than BHIB (ranged from 0.1907 to 2.0202 min−1), peach juice (ranged from 0.3733 to 1.5151 min−1) and milk (ranged from 0.2096 to 0.0.9477 min−1) at same pressure. The z values of L. monocytogenes were 578, 480, 506 and 576 for L. monocytogenes in milk, BHIB, peach juice and orange juice, respectively. There were significant (p<0.01) difference among kinetic values of AB and L. monocytogenes inactivation in broth and foods.

Physical and mechanical properties of pigeon pea, other grains and seeds are necessary for the design of equipment to handle, transport, process and store the crop. The physical properties of pigeon pea have been evaluated as a function of grain moisture content varying from 5% to 25% wet basis. In this seed moisture content range, the grain surface area, volume and sphericity decreased nonlinearly from 100.28 to 47.39 mm2, 94 to 28 mm3 and 0.91 to 0.82 respectively, while true density, bulk density and porosity increased nonlinearly from 0.75 to 1.22 g/mm3, 0.70 to 0.87 g/mm3 and 8% to 28% respectively. The 1000 seed mass, angle of repose and terminal velocity increased linearly from 96 to 140 g, 17° to 31° and 6.52 to 11.58 m/s respectively. The coefficient of static friction increased nonlinearly from 0.28 to 0.51 for plywood, 0.23 to 0.38 for galvanised steel and 0.18 to 0.31 for aluminium.

Two series of mayonnaise samples were prepared and used as model mayonnaise to study the slip phenomenon during shear measurements. The slip phenomenon, which is due to the non-homogenous stress field and/or direct fluid-geometry surface interactions, was examined by setting different gaps in a parallel disk rheometer. The magnitude of the slip velocity was greatly affected by the oil concentration and/or xanthan gum concentration as well as the applied shear stress during the shear measurement. The true shear rate could be corrected with measured apparent shear rate using a simple model, but the applied shear stress was limited due to the weak structure of the mayonnaise sample and measuring range limits of plate-plate geometry.

One of the challenges of food system development during a long term space mission is the need for reheating of food, processing and stabilization of waste products, including food and biological waste. We have previously reported on the development of a pouch with electrodes, which permits reheating and sterilization of its internal contents, and the development of a two-dimensional heat transfer model for design optimization. However, a fully three-dimensional model is necessary for ensuring that thermal processes result in sterility of all parts of the product, particularly edge regions. The ohmic heating of tomato soup in a pouch was simulated using computational fluid dynamics (CFD) codes with user defined functions (UDFs) for electric field equations. In general, good agreement was observed between model and experiment, excepting a zone at the edge, where the model tended to underpredict temperature. The 3D model permitted identification of the potential cold spots over the entire pouch, which was not obvious using our previous 2D model. In particular, the bottom corners of the pouch were found to be a zone of low current density, and showed a temperature of only 53.3 °C even when the peak pouch temperature was 139 °C. The 3D model is a useful tool to optimize electrode configurations, and to assure adequate sterilization processes.

Transient temperature and fluid flow during natural convection heating of a cylindrical can containing large food particles is simulated. The three dimensional equations of mass, momentum and energy conservation are solved using the finite volume method.A can (11 cm high and 10 cm in diameter) containing nine spherical particles (45 mm in diameter) is considered. Two meshes are tested. The first uses 1,650,000 tetrahedral cells. The second uses only 95,000 cells but with smaller prismatic cells near the can walls in order to better take into account the boundary layer phenomena. The simulated temperature evolutions in the liquid and at the centre of the particles obtained with the second grid are in good agreement with the experimental results. The liquid flows upwards in a thin boundary layer (about 2 mm thick) and flows downwards in the interstice between the particles. Due to thermal stratification, the slowest heating is for the particles at the bottom of the can.

The flow and viscoelastic properties of mayonnaise at different oil and xanthan gum concentrations (75–85% and 0.5–1.5% (w/w), respectively) were investigated in the rotational and oscillatory mode using a plate-plate rheometer. Yield stress, which was determined using a static method, and steady measurements were corrected to account for slippage. The corrected flow curves were fitted with the Herschel-Bulkley model, and it was found that the flow index (n), consistency index (K), and yield stress were greatly affected by the oil and xanthan gum concentrations. Viscoelastic properties of mayonnaise were characterized using small amplitude oscillatory shear, and it was observed that mayonnaise exhibited weak gel-like properties. The gel strength depends on the oil and xanthan gum concentrations. The magnitude of elastic modulus and complex viscosity increased with the increase of oil or xanthan gum concentrations.

The growth and bacteriocin production by Lactococcus lactis subsp. lactis CECT 539 was studied in batch and in re-alkalized fed-batch fermentations in media prepared with whey and mussel-processing wastes. From these cultures, mathematical models were developed to describe the growth and nisin production. The growth model developed which was based on Monod kinetics, has clearly more mechanistic approach and provides a major biological interpretability of the parameters than the logistic growth equation. This is the first study presenting a pseudomechanistic model to describe the growth kinetics of lactic acid bacteria in re-alkalized fed-batch fermentations. Nisin production was modeled with a modified form of the Luedeking and Piret model, which includes a term for the effect of the optimal final pH value in bacteriocin synthesis. Both models were demonstrated for three fed-batch cultivations of L. lactis using different culture media, feeding substrates and times of re-alkalization and feeding.

Thermal conductivity (k) of selected liquid foods during high- pressure processing (HPP) was studied using a line heat source probe. The probe was calibrated using distilled water and probe specific calibration factors were developed by comparing experimental data against published data from National Institute of Standards and Technology (NIST) for water. k of commercially available apple juice, canola oil, clarified butter, honey and high fructose corn syrup (HFCS) were then determined using a custom made high pressure experimental setup for various pressures (0.1, 100, 300, 500 and 700 MPa) at 25 °C. Results indicated that material k increased linearly with increasing pressures up to 700 MPa. Water and water-like substances (apple juice) were found to have the highest k values (up to 0.82 W/m °C at 700 MPa), while fatty foods such as canola oil and clarified butter had the lowest (0.29–0.4 W/m °C, respectively at 700 MPa) values. Honey and HFCS had intermediate values. The combined uncertainty (including Type A and Type B) in the measurement of k values of various liquid foods ranged from 0.6% (canola oil) to 3.8% (HFCS).

Spirulina platensis, a blue-green algae, is a potential source of the nutraceutical, γ-linolenic acid (GLA). The present work reports on recovery of GLA from S. platensis ARM 740 by supercritical carbon dioxide extraction (SCE) in comparison with conventional solvent extraction. Response Surface Methodology (RSM) was applied to optimize the operating conditions for high recovery of GLA by SCE. The levels studied were a pressure range between 100 and 500 bars, a time period between 26 min and 94 min and an ethanol level of 9.64–16.4 ml ethanol/16 g of freeze-dried biomass. The use of ethanol as a co-solvent with CO2 considerably increased the GLA yields compared to SCE. A recovery of 102% GLA in the supercritical extract (as compared to Bligh and Dyer extraction) was obtained in 1 h using a minimum of 13.7 ml of ethanol as co-solvent per 16 g of biomass, a temperature of 40 °C and a pressure of 400 bars. Supercritical CO2 with a co-solvent is therefore recommended as a better option to the conventional solvents for the complete recovery of GLA from S. platensis.

Electrical conductivity of select liquid foods and salt solutions was measured in situ during high pressure processing using a specially designed parallel electrode conductivity cell. Cell constants at atmospheric pressure were determined with KCl standards and calculated against standard data, while cell constants under pressure were estimated assuming isotropic compression. Measured conductivities of NaCl solutions under pressure were within 5.7% of previously reported data at pressures up to 800 MPa and temperatures to 61 °C. Electrical conductivity of NaCl and KCl solutions, orange juice, apple juice, tomato juice, and soybean oil were measured in triplicate in 100 MPa increments from 0.1 to 800 MPa. 0.01 m salt solutions were measured at 25 and 50 °C; 0.1 m salt solutions, juice and oil samples were measured at 25 °C. Results show conductivity of salt solutions and juice samples increased as a function of pressure, peaking between 200 and 500 MPa and decreasing above 500 MPa. Except for soybean oil, pressure had a significant effect (p < 0.01) on electrical conductivity for all samples. Temperature had a significant effect (p < 0.01) on electrical conductivity of 0.01 m salt solutions at all pressures. Conductivity of soybean oil was too low to be measured at atmospheric and pressurized conditions.

Microwave heating of prepared meals in sealed retail containers can be used to extend their shelf life. A numerical model to predict food temperatures after heating was developed and tested for use in the design of microwave cavities and foods. The model uses a finite difference scheme to predict temperatures by solving Maxwell's equations for electromagnetic fields and the heat conduction equation.Complex-shaped trays containing mashed potato were heated in a microwave tunnel operating at 2.3 kW for 60 s. The temperature distribution in the food was then measured over a 120 s period, during which the food partially cooled. The heat capacity and dielectric properties of the potato, and the power dissipated in the food during heating, were measured. Intricate temperature profiles, with large temperature gradients, were measured and predicted. The model predicted the positions of regions of highest and lowest temperature but large differences of up to 30 °C were found between measurements and predictions. The differences were largest near to regions where the finite difference mesh, which consisted of 4 mm cubes, was least able to simulate accurately the shape of the tray. Predictions carried out on a high performance workstation required 43 h of cpu. Further studies using a finer mesh on a more powerful computer, along with experiments using simple regular shapes of food, are required to test the model further.

Collaborative measurements of the thermal properties of foods and model foods are reported. They were part of a European Cooperation in Science and Technology (COST) project and provide data on thermal diffusivity and thermal conductivity of fish paste, meat paste, yoghurt, milk powder, apple pulp. ‘Model’ foods also examined were various mixtures of carrageenan, sugar and water and a monodisperse ‘powder’ of glass beads. Dependence of the measured properties on temperature, density and composition was also examined.

The drying behaviour, shrinkage and moisture distribution within cylindrical pieces of plantain, of varying thickness, and with different air temperatures, were studied in an experimental hot-air drier. By means of response surface methodology the influence of air temperature and thickness of the pieces on the drying rates was established. Air temperature had the greatest influence on drying behaviour. The activation energy for air-drying of plantain was estimated as 38.81 kJ (g.mol)−1. Shrinkage during drying was followed by changes in volume and dimensions of the pieces. Change in volume could be well described by a core drying model, while change in dimensions was related linearly with moisture content. A simulation, based on Fick's diffusion equation, was used to predict the distribution of moisture within the plantain pieces during drying. The predicted values related well to those obtained experimentally at moisture contents less than 1.20kg (kg d.s.)−1. This type of model could be used to determine local moisture contents rather than average moisture contents in food pieces, thus predicting more accurately the susceptibility of the food to spoilage.

Material dehydrated in sucrose solution by osmosis for 3 h at 30 °C and for 1 h at 70 °C was stored for 24, 72 and 144 h at 5, 9, 15 and 20 °C. Internal gradients of dry matter and sucrose concentration caused mass transfer in the material. Dry matter content in the surface layers was close to 40%, while sucrose concentration was between 20% and 23%. At the distance of 10 mm from the mass transfer surface dry matter content was still higher than that in raw apple, but sucrose concentration was that of raw apple. Concentration profiles of sucrose changed with time and storage temperature. The higher was the temperature the faster was the diffusion of sucrose. In material osmosed at 30 °C for 3 h and stored for 72 h at 5, 9 and 15 °C sucrose concentration gradients were still present. However, after 144 h at 9 °C the concentration of sucrose became independent on the distance from the mass exchange surface. Diffusion of sucrose in material osmosed at 70 °C for 1 h was much faster than that observed in samples osmosed at lower temperature. The effective diffusion coefficient of sucrose was dependent on temperature and time of storage, and distance from the mass transfer surface as well. Since distance was related to concentration a relationship between sucrose concentration and effective diffusion coefficient was found. The effective diffusion coefficient was of the order of 10−9–10−12 m2/s.

The moisture diffusion during freezing of porous medium has been widely studied for low porosity matrix (i.e. concrete, rocks, …), whereas very few works are available for materials with high porosity. This work focuses on the evaluation of moisture diffusion of a model food (sponge) with a high porosity (in the range of porosity 0.90–0.94). The drying data has been obtained at positive temperatures (10–30 °C) for selected water content (25–142 g H2O/100 g DM). Then, the effective diffusivities have been estimated by using two different methods. Finally, moisture diffusivity in the sub-zero domain has been extrapolated by using an Arrhénius’s model. During drying, the diffusivity increases as the average moisture content decreases. The temperature plays a significant role on the diffusivity. One of the two methods gives better results for the estimation of the effective diffusivity.

This study involves designing a branch shaker operating with an air compressor and its application for harvesting olives. Fruit detachment force (FDF) and fruit weight (W) were measured and hence the changes in the percentage of removed fruits with shaking were determined at different maturity times. During the harvesting season, two sets of experiments were conducted to test the shaker in the removal of Halhali cultivar olives (Olea europaea L.) with and without application of abscission chemical (ethrel). Experiments were also carried out to determine the effect of the chemical on fruit detachment force, fruit weight, and mechanical harvesting efficiency.Shaking tests showed that the removal percentage of olives without chemical application was lower than 50%. However, application of abscission chemical reduced the FDF/W ratio and thus increased the removal percentage of olives. The lowest fruit detachment force and the highest fruit removal (96%) were obtained by employing a frequency of 24 Hz and 12.5 mL/L concentration of abscission chemical with constant amplitude of 60 mm. Moreover, it was found that the fruit removal efficiency increases with decreasing olive detachment forces.

Dynamic changes in absolute residual PME activity in potato tubers undergoing blanching were analysed with regard to the temperature distribution in the tissue after the treatment. The finite-element technique was used to predict the temperature distribution within the tissue during blanching. The temperature-related behaviour of the enzyme indicated that blanching between 60 and 70 °C induced thermal activation of PME in blanched potatoes. The temperature distribution in the tissue during blanching showed that activation takes place as soon as the tissue reaches an average temperature above 50 °C. However, the temperature dependence of the absolute residual PME activity in the tissue, when blanching at temperatures between 50 and 70 °C, did not follow Arrehenius' law. The degree of PME activity in the blanched tissue was found to depend on the “effective exposure time” to the temperature at which the activation/deactivation of the enzyme takes places. Thermal activation of PME in potato tissue occurred during the first two minutes of blanching. Longer heating times induced rapid deactivation of the enzyme in the tissue.

A transient analytical model has been presented to study the new concept of a solar crop dryer having reversed absorber plate type collector and thermal storage with natural airflow. The performance of 1 × 1 m2 area of crop dryer with packed bed and airflow channel was evaluated for drying of onions. The model was solved to compute the air temperatures and various functional components of the drying systems for a day of the month of October for the climatic condition of Delhi (India). The parametric study involved the effect of width of airflow channel and height of packed bed on the crop temperature. The effect of thermal storage is observed on the natural mass flow rate in the drying system. A 30° inclined absorber plate with in-built thermal storage and 0.12 m width of airflow channel induced the mass flow rate in the range of 0.032–0.046 kg s−1 during the drying process. The thin layer drying equation was used to study the drying rate and hourly reduction in moisture content in the crop trays. It has been observed that the crop moisture content and drying rate decreases with the drying time of the day. A reversed absorber plate of 1 m length and 1 m breadth with 0.15 m packed bed could dry 95 kg of onion from a moisture content of 6.14–0.27 kg water/kg of dry matter in a 24 h drying period.

Absorption kinetics of two commercial O2 and CO2 scavengers commonly used in active modified atmosphere packaging (MAP), were studied. Individual scavenger sachets were placed in polyvinylidene chloride pouches filled with air or modified atmosphere at 0% or 100% relative humidity and at 5, 20 and 35 °C. The headspace gas composition was measured as a function of time. Absorption kinetics were described by a first-order reaction with an Arrhenius type behaviour. The absorption capacity, absorption rate constant, energy of activation, Arrhenius constant and variation of all these parameters were evaluated and discussed. Significant “parasite” CO2 absorption was observed for O2 scavengers. This study also underlined the need to take into account the important variation of absorption rate constant (about 20%) among individual gas scavengers and the temperature effect for reliable evaluation of the gas kinetics when using O2 or CO2 scavengers in an expected and secure way.

Theoretical and experimental studies of absorption-driven multiple-effect evaporators were carried out. Theoretical predictions for such units were compared with existing industrial-scale conventional evaporators. It was calculated that the total energy consumption of an absorption-driven evaporator is considerably lower than that of a conventional multiple-effect evaporator. Theoretical predictions were also verified with an experimental four-effect absorption-driven evaporator, coupled with a two-effect regenerator, using sodium hydroxide solution as the absorptive medium. Energy savings of 30% to 45% were obtained.

Water absorption during soaking of maize, millet and sorghum at 10°C, 30°C and 50°C was studied using Peleg's equation. The equation gave a reasonable fit to experimental data. Peleg constants were obtained for the cereals. The constant K2 was unaffected by temperature of soaking. Temperature dependence of the reciprocal of the Peleg constant K1 was determined using an Arrhenius equation. Activation energy was in the range 13·99–16·23 MJ mol−1 compared to 19·02–19·56 MJ mol−1 obtained for soybean, cowpea and undehulled groundnut. An exponential relationship was proposed to describe the relationship between the rate of absorbed water per unit change in temperature and the activation energy.

An apparatus to determine the solubility and rate of absorption of CO2 from pressure changes over time was built and validated against water. A high correlation (r2>0.99) between the measured solubility and theoretical solubility was found. The solubility determinations using the manometric method were more accurate than using a CO2 electrode. The solubility of CO2 increased linearly with the partial pressure of CO2, and a quadratic response was observed for the gas volume/product volume (g/p) ratio. The temperature dependency of the solubility is according to the temperature dependency of Henry’s constant. The rate of absorption is primarily affected by the %-level of CO2 in the atmosphere, and the effective diffusion constant is primarily affected by the %-level of CO2 and the g/p-ratio. The time to reach 50% solubility was about 3 h and equilibrium was reached after 48 h.

The solubility and diffusion of carbon dioxide into salmon, cod, anglerfish, wolf-fish and tuna fillets was determined by monitoring pressure changes over time in a closed chamber at constant temperatures. High correlation (r2>0.95) between the measured solubility against the theoretical solubility of water corrected for the total water and fat content of the fish was found. The average Henry’s constant for raw fish fillets at 0 °C was calculated to be 45.1 Pa ppm−1 (mg CO2 per kg fish).Using the un-steady state solution of Fick’s law for the pressure history from 0 to 15 h gave better measurements of the CO2 solubility in fishes than using the pressure drop over 20 h, mainly because the system does not reach equilibrium until about three days. In addition this graphical solution made determination of the diffusion coefficient possible. The average effective diffusion coefficient for CO2 into raw fish fillets at chilled and superchilled temperatures was determined to be 1.69 × 10−5 cm2/s.

The effects of temperature, drying method and pre-drying treatments on water absorption and textural degradation of potatoes during soaking were examined. Samples were pre-dried in a convective oven (60 °C) or in a microwave oven (250, 440 or 600 W), and subsequently rehydrated in a water bath at temperatures between 20 and 80 °C. The texture of potatoes (firmness) was measured using a texture analyser. Fick’s Second Law of Diffusion was used to describe the rehydration kinetics (diffusion coefficient, D and equilibrium moisture content, Xe). Textural degradation kinetics were estimated using the fractional conversion equation. The rehydration characteristics and textural degradation kinetics were dependent on temperature, pre-blanching, drying method (convective or microwave) and pre-soaking (ionic surfactants and NaCl).

Cross mass transfer including water entrance and solute loss from Rice Crispies were characterised and underlying mechanisms are proposed. Various techniques were tested, and conventional techniques involving phase separation, weighing and oven-drying for determination of water content showed wide variability due to the delicate operation of phase separation. A non-destructive, non-invasive NMR method was also developed. Special care was paid to the contribution of non-aqueous protons to the liquid signal from hydrated Rice Crispies. The kinetics of water absorption obtained from NMR relaxation amplitudes were within the range of those obtained from conventional techniques. It was also possible to monitor solute leaching from the relaxation rates of protons from the immersion water.

Rectangular prisms of chestnuts (dimensions 10 × 10 × 15 mm) previously dried at 65 °C were rehydrated at different moisture contents (0.15, 0.24, 0.33, and 0.66 kg water/kg d.b.) by immersion in water during different periods of time (up to 180 min) and temperatures (25, 45, 70, and 100 °C). Rehydration kinetics were monitored by measuring their weights at regular intervals and were characterized using several rehydration indexes such as dry basis holding capacity (DHC), water absorption capacity (WAC), and rehydration ability (RA). Simultaneously, other properties that determine quality such as volume recovery, texture (by means of compression tests), and color during rehydration were experimentally determined.Peleg’s model was successfully applied to experimental data and the corresponding parameters were obtained and correlated with temperature. Leaching flow was important only at high temperatures and dried samples at low moisture content gave satisfactory results for rehydration indexes. Texture kinetics was satisfactorily described by means of second-order kinetic model and the corresponding constant rate showed an Arrhenius relationship with temperature. Color was not recovered during rehydration. The results indicate that the use of low temperatures is more than adequate for preserving this property. Globally, during rehydration, chestnut showed acceptable rehydration rates taking into account the low porosity of the product, but rehydrated samples were softer and darker than fresh samples. Low-temperature rehydration and high initial moisture content of sample give better global properties for the rehydrated product.

Water absorption characteristics of paddy, brown rice and husk were measured at three temperatures ranging from 30 to 60 °C. From the water absorption characteristics curve, it was observed that the husk was a significant barrier in the water absorption process by brown rice. Using the measured moisture data, a non-linear regression method was applied to an approximate solution of the diffusion equation MR = A1exp(−kt) for an infinite cylinder shape. The geometrical shape factor was estimated using the value of constant A1 and the characteristics length. The predicted value of moisture content at any time was in good agreement with the observed data. The mean values determined for the diffusion coefficients were 4.91 × 10−11 m2/s for paddy, 9.56 × 10−11 m2/s for brown rice and 1.16 × 10−08 m2/s for husk. Analysis of variance showed that soaking temperatures did not have significant effect on diffusion coefficients. An Arrhenius-type equation was used to relate the diffusion coefficient of paddy, brown rice and husk to absolute temperature (K) and the energy of activation was estimated. The values determined were 31.50 kJ/mole for paddy, 37.32 kJ/mole for brown rice and 19.25 kJ/mole for husk.