Batch Liquid-Liquid Extraction of Phenol from Aqueous Solutions

The aim of this work is the study of batch liquid-liquid extraction of phenol from aqueous solutions in a bench-scale well-mixed reactor. The influence of the ratio of phase volumes, temperature, and rotational speed on phenol removal (0.72-1.1% w/w) was investigated using methyl isobutyl ketone as an extracting solvent. For this purpose, the ratio of phase volumes were set at 0.1 and 0.2, the temperature at 10, 20, and 30 degrees C, and the rotational speed at 300, 400, and 500 rpm. A physical model based on the material balance of the phases as well as the equation of mass flux between the phases allowed the estimation of the overall coefficient of mass transfer coupled with the superficial area. Moreover, it proved to fit, satisfactorily well, the experimental data of residual phenol concentration in the organic phase versus time under all the conditions investigated.

Microencapsulation of casein hydrolysate by complex coacervation with SPI/pectin

The aim of this work was to encapsulate casein hydrolysate by complex coacervation with soybean protein isolate (SPI)/pectin. Three treatments were studied with wall material to core ratio of 1:1, 1:2 and 1:3. The samples were evaluated for morphological characteristics, moisture, hygroscopicity, solubility, hydrophobicity, surface tension, encapsulation efficiency and bitter taste with a trained sensory panel using a paired comparison test. The samples were very stable in cold water. The hydrophobicity decreased inversely with the hydrolysate content in the microcapsule. Encapsulated samples had lower hygroscopicity values than free hydrolysate. The encapsulation efficiency varied from 91.62% to 78.8%. Encapsulated samples had similar surface tension, higher values than free hydrolysate. The results of the sensory panel test considering the encapsulated samples less bitter (P < 0.05) than the free hydroly-state, showed that complex coacervation with SPI/pectin as wall material was an efficient method for microencapsulation and attenuation of the bitter taste of the hydrolysate. (C) 2009 Elsevier Ltd. All rights reserved.

Production and properties of casein hydrolysate microencapsulated by spray drying with soybean protein isolate

The aim of this work was to encapsulate casein hydrolysate by spray drying with soybean protein isolate (SPI) as wall material to attenuate the bitter taste of that product. Two treatments were prepared: both with 12 g/100 g solids and containing either two proportions of SPI: hydrolysate (70:30 and 80:20), called M1 and M2, respectively. The samples were evaluated for morphological characteristics (SEM), particle size, hygroscopicity, solubility, hydrophobicity, thermal behavior and bitter taste with a trained sensory panel using a paired-comparison test (non-encapsulated samples vs. encapsulated samples). Microcapsules had a continuous wall, many concavities, and no porosity. Treatments M1 and M2 presented average particle sizes of 11.32 and 9.18 mu m, respectively. The wall material and/or the microencapsulation raised the hygroscopicity of the hydrolysate since the free hydrolysate had hygroscopicity of 53 g of water/100 g of solids and M1 and M2 had 106.99 and 102.19 g of water/100 g of solids, respectively. However, the hydrophobicity decreases, the absence of a peak in encapsulated hydrolysates, and the results of the panel sensory test considering the encapsulated samples less bitter (p < 0.05) than the non-encapsulated, showed that spray drying with SPI was an efficient method for microencapsulation and attenuation of the bitter taste of the casein hydrolysate. (c) 2008 Elsevier Ltd. All rights reserved.

The influence of ARIMA-GARCH parameters in feed forward neural networks prediction

The objective of this article is to find out the influence of the parameters of the ARIMA-GARCH models in the prediction of artificial neural networks (ANN) of the feed forward type, trained with the Levenberg-Marquardt algorithm, through Monte Carlo simulations. The paper presents a study of the relationship between ANN performance and ARIMA-GARCH model parameters, i.e. the fact that depending on the stationarity and other parameters of the time series, the ANN structure should be selected differently. Neural networks have been widely used to predict time series and their capacity for dealing with non-linearities is a normally outstanding advantage. However, the values of the parameters of the models of generalized autoregressive conditional heteroscedasticity have an influence on ANN prediction performance. The combination of the values of the GARCH parameters with the ARIMA autoregressive terms also implies in ANN performance variation. Combining the parameters of the ARIMA-GARCH models and changing the ANN`s topologies, we used the Theil inequality coefficient to measure the prediction of the feed forward ANN.