Artificial neural network modeling for temperature and moisture content prediction in tomato slices undergoing microwave-vacuum drying

ABSTRACT:  Inputs for ANN (multihidden-layer feed-forward artificial neural network) models were drying time ( t _{i + 1}), initial temperature ( T ₀), moisture content ( MC ₀), microwave power, and vacuum pressure. The outputs were temperature ( T _{i + 1}) and moisture content ( MC _{i + 1}) at a given t _{i + 1}. After training the ANN models with experimental data using the Levenberg-Marquardt algorithm, a two-hidden-layer model (25-25) was determined to be the most appropriate model. The mean relative error (MRE) and mean absolute error (MAE) of this model for T _{i + 1} were 1.53% and 0.77 °C, respectively. In the case of MC _{i + 1}, the MRE and MAE were 11.48% and 0.04 kg_water/kg_dry, respectively. Using temperature ( T _i ) and moisture content ( MC _i ) values at t _i in the input layer significantly reduced the computation errors such that MRE and MAE for T _{i + 1} were 0.35% and 0.18 °C, respectively. In contrast, these error values for MC _{i + 1} were 1.78% (MRE) and 0.01 kg_water/kg_dry (MAE). These results indicate that ANN models were able to recognize relationships between process parameters and product conditions. The model may provide information regarding microwave power and vacuum pressure to prevent thermal damage and improve drying efficiencies.     

Influence of vitamin E supplementation on spontaneous oxidized flavour of milk in dairy buffaloes

Twenty-four Murrah buffaloes (60 days pre-partum) were divided into four equal groups (T ₁ , T ₂ , T ₃ and T ₄ ) and were supplemented with 0, 1000, 1500 and 2000 IU α-tocopheryl acetate per day up to 30 days of lactation, and half of these doses from 30 to 60 days of lactation. Milk samples collected fortnightly were analysed for vitamin E, fat, and development of oxidized flavour, with and without copper addition by a panel of judges, and chemically by the thiobarbituric acid test. Scores for oxidized flavour ranged from 0 to 10 with 0–4 as definite, 5–7 as light and 8–10 having no defect. The α-tocopherol content in milk fat (µg/g) averaged 20.55, 25.56, 29.98 and 31.38 in T ₁ , T ₂ , T ₃ and T ₄ groups, respectively. The addition of Cu in the milk significantly increased milk fat oxidation. Better stability of milk in T ₃ and T ₄ groups was observed, which might be due to a higher level of milk α-tocopherol. Addition of 1500 IU α-tocopheryl acetate in the diet of buffaloes helped in improving the oxidative stability of milk.     

Artificial neural network for the prediction of temperature, moisture and fat contents in meatballs during deep-fat frying

An artificial neural network (ANN) was developed to predict heat and mass transfer during deep-fat frying of meatballs. Frying time, radius of meatball, fat diffusivity, moisture diffusivity, heat transfer coefficient, fat conductivity, initial moisture content, thermal diffusivity, initial meatball temperature and oil temperature were all input variables. Temperature at meatball geometrical centre ( T ₀), average temperature of meatball ( T _ave), average fat content of meatball ( m _f,ave), and average moisture content of meatball ( m _ave) were outputs. The data used to train and verify the ANN were obtained from validated mathematical models. Trained ANN predicted T ₀, T _ave, m _f,ave and m _ave with 0.54, 0.14, 0.03 and 0.10% mean relative errors, respectively.     

Confidence intervals for modeling anthocyanin retention in grape pomace during nonisothermal heating

ABSTRACT:  Degradation of nutraceuticals in low- and intermediate-moisture foods heated at high temperature (>100 °C) is difficult to model because of the nonisothermal condition. Isothermal experiments above 100 °C are difficult to design because they require high pressure and small sample size in sealed containers. Therefore, a nonisothermal method was developed to estimate the thermal degradation kinetic parameter of nutraceuticals and determine the confidence intervals for the parameters and the predicted Y (concentration). Grape pomace at 42% moisture content (wb) was heated in sealed 202 × 214 steel cans in a steam retort at 126.7 °C for > 30 min. Can center temperature was measured by thermocouple and predicted using Comsol software. Thermal conductivity (k) and specific heat ( C_p ) were estimated as quadratic functions of temperature using Comsol and nonlinear regression. The k and C_p functions were then used to predict temperature inside the grape pomace during retorting. Similar heating experiments were run at different time–temperature treatments from 8 to 25 min for kinetic parameter estimation. Anthocyanin concentration in the grape pomace was measured using HPLC. Degradation rate constant ( k _{110 °C}) and activation energy ( E_a ) were estimated using nonlinear regression. The thermophysical properties estimates at 100 °C were k = 0.501 W/m °C, Cp = 3600 J/kg and the kinetic parameters were k _{110 °C}= 0.0607/min and E_a = 65.32 kJ/mol. The 95% confidence intervals for the parameters and the confidence bands and prediction bands for anthocyanin retention were plotted. These methods are useful for thermal processing design for nutraceutical products.     

Effect of pan-frying in extra-virgin olive oil on odour profile, volatile compounds and vitamins

Changes in odour of Arauco (ARA) and Arbequina (ARB) extra-virgin olive oil (OO) were monitored during frying by electronic nose (EN) and solid-phase microextraction–gas chromatography methodologies. Degradation of α- and γ-tocopherols was monitored by HPLC. Electronic nose data and volatile compounds were analysed at intervals of 15 min ( t ₁₅) during 60 min of frying ( t ₆₀). α- and γ-tocopherols were determined at intervals of 5 min ( t ₅) during 30 min of frying ( t ₃₀). Principal components analysis applied to EN data showed one component, PC₁ which accounted 96.6% of the total odour variation. SnO₂ sensors had a positive correlation with PC₁. ARA variety corresponding to frying t ₆₀ had the highest positive correlation with PC_1. Analysis of variance results for volatile compounds showed an increase on production for: 3-methyl butanal, n -pentanal, n -hexanal, n -heptanal and n -nonanal at 15 min of frying for ARB OO and at 30 min for ARA OO. α-tocopherol and γ-tocopherol showed an important decrease after the first 5 min of frying for ARB OO and at 15 min for ARA OO.     

Statistical and Kinetic Studies of the Changes in Soybean Quality during Storage as Related to Soymilk and Tofu Making

ABSTRACT:  The objective of this study was to develop statistical equations and kinetic models to describe the changes of soybean quality during storage. Significant correlations ( P  < 0.0001) were found among most of quality attributes including color parameters (Hunter L , a , b , and Δ E ), solid extractability (as expressed by soymilk solids content), soymilk pH and protein content, tofu yield, hardness, and protein content. Regressed linear equations were developed between color indices ( L/L ₀, ΔE ) and soymilk/tofu making properties. Empirical equations were developed to relate soybean color indices ( L/L ₀, ΔE ) and storage conditions including variables of initial moisture content (MC), relative humidity (RH), temperature ( T ), and duration ( t ). Kinetics of the changes in soybean color and extractability during storage at 70% RH and 22 to 40 °C were investigated. The kinetics was well described by zero-order kinetics. The Arrhenius equation adequately described the temperature dependence of the reaction rate constants for all parameters, from which the activation energies and rate constant were obtained. The equations developed in this study provided simple methods to monitor soybean quality and predict quality changes of soybeans during storage at various conditions.     

Comparison of the Dispersed Phase Coalescence Mechanisms in Different Tablespreads

ABSTRACT:  Temperature-induced destabilization of the dispersed phase in butter and margarine was compared by following changes in droplet size ( d _3,3), solid fat content (SFC), and fat crystal spatial organization in the 28–34 °C range. At 28 °C, both butter and margarine were stable, with similar d _3,3 values (approximately 6 μm) and droplet size distributions. As the storage temperature was raised above 30 °C, notable droplet coalescence was observed (for example, at 32 °C d _3,3 values of approximately 10 μm for butter and approximately 12 μm for margarine were obtained). Dispersed phase coalescence in butter was dominated by coagulation, with the fat crystal network-limiting droplet–droplet contact until a minimum SFC was reached (approximately 2.5%). In margarine, the rate-limiting step for coalescence was the melting of Pickering crystals present around the dispersed aqueous droplets. Unlike butter, there was no sharp change in stability at a particular temperature or critical SFC. With these differences, coalescence in butter could be modeled as a 2nd-order process and as a 1st-order process in margarine. Overall, these results demonstrated that the kinetic stability of the dispersed aqueous phase in butter and margarine depends on SFC and the spatial distribution of fat crystals within the spreads.     

The rheological properties of ketchup as a function of different hydrocolloids and temperature

The flow properties of ketchup were assessed upon addition of commonly used food thickeners: guar, xanthan and CMC gum at three different concentrations (0.5%, 0.75% and 1%) and four temperatures (25, 35, 45 and 55 °C). The ketchup without supplementation served as a control. All ketchup formulations exhibited non-Newtonian, pseudoplastic behaviour at all temperatures and hydrocolloid levels. The Power-law and Herschel-Buckley model were successfully applied to fit the shear stress versus shear rate data. The flow behaviour indices, n and n' , varied in the range of 0.189–0.228 and 0.216–0.263, respectively. The consistency coefficients, k and k' , were in the range of 8.42–27.22 and 6.56–20.10 Pa s ⁿ , respectively. The addition of hydrocolloids increased the yield point (τ₀) and apparent viscosity of the ketchup in comparison to that of the control. The Arrhenius equation was successfully used to describe the effects of temperature on the apparent viscosity of the prepared formulations. The E _a value appeared in the range between 5492.6 and 21475.8 J mol⁻¹.     

Development of a simple model based on chemical kinetics parameters for predicting respiration rate of carambola fruit

The respiration model was parameterised with experimental data obtained at temperatures 10, 15, 20, 25 and 30 °C by the closed system. Reaction orders of O₂ and CO₂ determined for respiration rate are 0.912 and −0.24, respectively. Estimates of A and E _a for rate coefficient are 5.219 _× 10¹⁰ h⁻¹ and 61.397 kJ mol⁻¹, respectively. The reliability of the model was verified against new data generated from 18 °C. An enzyme kinetics based model was likewise built and verified as a contrast. Verification results indicate that the chemical kinetics model performs very well and has superiority over the enzyme kinetics one. This study demonstrates that chemical kinetics can be used to underpin the development of simple respiration model and should facilitate the prediction of respiration rate for the carambola fruit in practice.     

Inverse Method to Estimate Kinetic Degradation Parameters of Grape Anthocyanins in Wheat Flour Under Simultaneously Changing Temperature and Moisture

ABSTRACT:  Thermal and moisture effects on grape anthocyanin degradation were investigated using solid media to simulate processing at temperatures above 100 °C. Grape pomace (anthocyanin source) mixed with wheat pastry flour (1: 3, w/w dry basis) was used in both isothermal and nonisothermal experiments by heating the same mixture at 43% (db) initial moisture in steel cells in an oil bath at 80, 105, and 145 °C. To determine the effect of moisture on anthocyanin degradation, the grape pomace–wheat flour mixture was heated isothermally at 80 °C at constant moisture contents of 10%, 20%, and 43% (db). Anthocyanin degradation followed a pseudo first-order reaction with moisture. Anthocyanins degraded more rapidly with increasing temperature and moisture. The effects of temperature and moisture on the rate constant were modeled according to the Arrhenius and an exponential relationship, respectively. The nonisothermal reaction rate constant and activation energy (mean ± standard error) were k _{80 °C, 43% (db) moisture} = 2.81 × 10⁻⁴± 1.1 × 10⁻⁶ s⁻¹ and Δ E  = 75273 ± 197 J/g mol, respectively. The moisture parameter for the exponential model was 4.28 (dry basis moisture content)⁻¹. One possible application of this study is as a tool to predict the loss of anthocyanins in nutraceutical products containing grape pomace. For example, if the process temperature history and moisture history in an extruded snack fortified with grape pomace is known, the percentage anthocyanin loss can be predicted.     

Determination of a process time for a new product: canned low acid artichoke hearts

In order to obtain a process time for canned low acid artichoke hearts, the heat resistance of Clostridium sporogenes PA 3679 in phosphate buffer (pH 7) and in artichoke puree (pH 5.2) was determined in the range of 100–118°C. D _T values in artichoke puree were determined by the most probable number and plate count methods. D _121°C values were deduced by extrapolation of the curves. An F 8.3/121 (see nomenclature section) target value of 1.8 min, equivalent to an F ₀= 2.5 min, was established considering a D _121°C value of 0.36 min for artichoke puree. The cold point for canned artichoke hearts in 1/2 kg (71.5x117 mm) cans was determined, a set of heat penetration and cooling curves were obtained and the values of the parameters ' j ' and ' f ' were found. A process time was deduced, 21.5 and 23 min of heating at 121°C, following the methods of Patashnik (1954) and Hayakawa (1970, 1974) respectively, in order to reach the F ₀ target of 2.5 min for this new canned food. This process was adequate, as the inoculated experimental pack test showed no altered cans out of 50 inoculated after 21.5 min of heating at 121°C.     

Texture of rehydrated dried bell peppers modified by low-temperature blanching and calcium addition

Diced green bell pepper was blanched twice, once at 51–79 °C for 19–61 min, and once at 95 °C for 3 min, and dried. The firmness of rehydrated samples was measured by puncture, and optimum conditions assessed by response surface methodology. The optimized model showed that, blanching at 65 °C for 49 min gave a 64% increase in puncture force over the control. The optimum temperature was used to evaluate the effect of adding CaCl₂. The dices were blanched twice, once at 65 °C for 3 min in either 0 or 4% CaCl₂, secondly in either 0 or 2% CaCl₂ solution at 95 °C for 3 min. In the second case the dices had been held at room temperature for 0–30 min before treatment. Adding CaCl₂ increased puncture force significantly ( P  ≤ 0.05). The best results, those which gave greatest firmness, were obtained by blanching at 65 °C for 3 min in 4% CaCl₂, holding for 16 min after blanching, followed by a secondary blanching at 95 °C in 2% CaCl₂.     

Changes in viscosity and pectolytic enzymes of tomato and strawberry juices processed by high-intensity pulsed electric fields

The effects pulse frequency (50–250 Hz), pulse width (1.0–7.0 μs) and polarity (monopolar or bipolar) of high-intensity pulsed electric field treatments (35 kV cm⁻¹ and 1000 μs) on viscosity and the pectin methylesterase (PME) and polygalacturonase (PG) activities were evaluated using a response surface methodology. Second-order expressions were accurate enough to fit experimental results. Tomato juice apparent viscosity increased within the range of the assayed conditions, achieving the highest values at 250 Hz and 7.0 μs in bipolar mode. At the same conditions the lowest residual PME (RA_PME = 10%) and PG (RA_PG = 45%) activities were observed in the juice. Apparent viscosity of strawberry juices slightly rose when frequencies higher than 100 Hz and 1-μs monopolar pulses were applied to the juice. Treatments causing the greatest increase in strawberry juice apparent viscosity also led to the lowest RA_PME (10%) and RA_PG (75%) values. In contrast, viscosity loss was promoted under the rest of assayed HIPEF conditions despite the low RA_PME values (<20%) achieved. Moreover, RA_PG did not decrease below 75% throughout the range of studied conditions.     

The use of spinning disc reactor for processing ice cream base – effect of ageing in making model ice cream

Spinning disc reactor (SDR) technology was tested to produce an ice cream base, which was subsequently used to make model ice cream. The ice cream base containing butterfat, lecithin, xanthan gum, sugar, skimmed milk and double cream was passed over the SDR disc spinning at 2900 rpm, heated at 80 °C and at a flow rate 6 mL s⁻¹. The physical properties of the SDR-processed ice cream base such as particle size and viscosity measurments, and of model ice cream including overrun, meltdown rate and sensory perception were investigated. The SDR-processed ice cream base exhibits narrow particle-size distribution (average particle d ₃₂ = 1.65 μm, d ₄₃ = 2.98 μm) and the viscosity was found to be similar at zero and 18 h ageing, whilst the model ice cream requires zero-hour ageing and has a high overrun value (∼85%) and slow meltdown rate as compared with a commercial sample. The results reveal that the SDR is capable of producing a highly stable ice cream base that requires significantly less ageing than the 18 h typically associated with the traditional process of making ice cream. The SDR process provides intense mixing of ingredients which facilitates the hydration of milk proteins and stabilisers.     

Rheological behaviour of acorn starch dispersions: effects of concentration and temperature

Rheological properties of acorn starch dispersions at different concentrations (4%, 5%, 6% and 7%) were evaluated under steady and dynamic shear conditions. The flow behaviours of the acorn starch dispersions at different temperatures (25, 40, 55 and 70 °C) were determined from the rheological parameters provided by the power law model. The acorn starch dispersions at 25 °C exhibited high shear-thinning fluid characteristics ( n  = 0.23–0.36). Consistency index (K) and apparent viscosity (η_a,100) increased with an increase in starch concentration, and were also reduced with increasing temperature. Within the temperature range of 25–70 °C, the η_a,100 obeyed the Arrhenius temperature relationship with a high determination coefficient ( R ² = 0.97–0.99), with activation energies (E_a) ranging between 16.5 and 19.0 kJ mol⁻¹. Both the power law and exponential type models were employed in order to establish the relationship between concentration and apparent viscosity (η_a,100) in the temperature range of 25–70 °C. Magnitudes of storage ( G' ) and loss ( G ") moduli increased with an increase in the starch concentration and frequency (ω). The magnitudes of G ' were higher than those of G " over most of the frequency range (0.63–62.8 rad s⁻¹). The dynamic (η*) and steady shear (η_a) viscosities of acorn starch dispersion at 7% concentration follow the Cox–Merz superposition rule.     

The effect of freezing and thawing on the magnetic resonance imaging parameters of water in beef, lamb and pork meat

A new multisample magnetic resonance imaging (MRI) technique has been used to measure the values of T ₁, T ₂, magnetization transfer (MT) rate and liquid proton density ratio ( M ₀%) for water in fresh and frozen–thawed beef, lamb and pork; four samples from the longissimus dorsi muscle of 25 different individuals per species were studied. The effect of two different chilling regimes, standard and tenderloin, were also studied. Standard chilled pork and lamb had a significantly higher M ₀% than tenderloin chilled samples, but there was no correlation between M ₀% and fresh weight; the differences in M ₀% are thought to be due to differences in MR (magnetic resonance)-visible water. There was a significant decrease in water content, T ₁ relaxation time, and a significant increase in MT rate in the frozen–thawed samples when all the data were pooled for each species. However, individual animals differed in the magnitude and direction of the change; this means that because of interanimal variation, measurements of MR parameters from a single sample of meat cannot be used at present to authenticate it.     

Gelation mechanism of surimi studied by 1h NMR relaxation measurements

ABSTRACT:  In order to elucidate the gelation mechanism of surimi, the temperature dependence of water proton spin–spin relaxation time (¹H T ₂) has been described by a theoretical approach, in which the exposed protein surface is taken into account. Water ¹H T ₂ measured for horse mackerel surimi in the presence of 2.5% NaCl was analyzed on the basis of the consideration for the denaturation and the aggregation of protein in order to explain the macroscopic structural change during the heating and the cooling processes. The temperature dependence of water ¹H T ₂ and the fraction of rigid component gave a clear explanation for the gelation mechanism of surimi. Differential scanning calorimetry thermogram and dynamic viscoelastic measurements supported the results of nuclear magnetic resonance (NMR) measurements. It has been demonstrated that the measurement of NMR relaxation times is useful to describe the gelation mechanism of surimi.     

Comparison of a computer evaluation with a standard evaluation of retort pouch thermal processing

Flexible retort pouches (188x127x25.4 mm) filled with 10% bentonite suspension, tomato ketchup and potato puree were processed under computer control using steam-air or hot water, and the heating/cooling curves and thermal diffusivities were determined. A computer method, developed for the evaluation of the thermal process using the one-dimensional heat conduction equation, was used to calculate F ₀ values which were then compared with those obtained using the General- and the Stumbo's Formula methods. Temperatures at the pouch centre estimated by the Formula- and the Computer method were in good agreement with those obtained experimentally. The F ₀ values estimated by the computer method were closer to the F ₀ values estimated by the General method than those obtained by the Formula method.     

Production, purification and thermal characterisation of invertase from a newly isolated Fusarium sp. under solid-state fermentation

Production of invertase employing a newly isolated Fusarium sp. under solid-state fermentation was optimised. Different process parameters were optimised. The maximum enzyme activity under optimum conditions was 47.23 ± 2.12 U gds⁻¹ with nitrogen additives. The enzyme was purified by ammonium sulphate precipitation, diethylaminoethyl cellulose ion-exchange chromatography and Sephadex gel filtration. This protocol gave 20.25-fold purification and 5.53% recovery. The optimum pH and temperature for activity were 5.0 and 50 °C. The K _m and V _max values for the enzyme were 8.33 m m and 21.48 μmol min⁻¹, respectively. A detailed kinetic study of thermal inactivation has been carried out. Enthalpy of activation (Δ H *) decreased when entropy (Δ S *) of activation increased at higher temperatures. Moreover, free energy of denaturation (Δ G *) increased at higher temperature making the enzyme thermally stable. A possible explanation for the thermal inactivation of invertase at higher temperatures is also discussed.     

Effect of different extraction and precipitation methods on yield and quality of pectin

An experimental study has been conducted to determine the combined effects of different extraction conditions and precipitation method on the yield and quality of high methoxyl pectin from lemon peels. Pectin was extracted using different mineral acids (HCl, HNO₃ and H₂SO₄) at four concentration levels (0.025; 0.05; 0.1 and 0.2  m ), at 70 °C for 4 h. The soluble pectin was precipitated by iso-propanol or by an aluminium sulphate, Al₂(SO₄)₃, solution at pH 4. The extraction with HCl and HNO₃, at the highest concentrations investigated, followed by aluminium precipitation led to the best results in terms of yield (22–25%), quality and gelling power of pectin with a remarkable decrease of alcohol consumption as compared to the alcoholic precipitation under the same extracting conditions.