Modeling and optimization of pectin extraction from banana peel using artificial neural networks (ANNs) and response surface methodology (RSM)

Extraction of phenolic acids and tetramethylpyrazine in Shanxi aged vinegar base on vortex-assisted liquid-liquid microextraction-hydrophobic deep eutectic solvent: COSMO-RS calculations and ANN-GA optimization

In this study, vortex-assisted liquid-liquid microextraction (VA-LLME) based on hydrophobic deep eutectic solvents (HDES) was used to efficiently and sustainably extract five phenolic acids and tetramethylpyrazine (TMP) from Shanxi aged vinegar (SAV). The VA-LLME technique was employed to investigate the extraction mechanism of HDES with the best extraction performance for the target compounds using a conductor-like screening model for real solvents (COSMO-RS). An artificial neural network combined with a genetic algorithm (ANN-GA) was developed to optimize the extraction conditions based on single-factor and response surface methodology, while also analyzing the interactive effects on the phenolic acids and TMP in the extracted solution during the extraction phase. The optimized conditions were determined, and the greenness of the procedure was evaluated using an analytical greenness metric, indicating that this technique can serve as a green alternative for the determination of phenolic acids and TMP in SAV.

Influence of ripening stage on the microwave-assisted pectin extraction from banana peels: A feasibility study targeting both the Homogalacturonan and Rhamnogalacturonan-I region

This work investigated a sustainable and efficient approach of pectin extraction for banana peel waste valorisation and studied the influence of banana ripening stages (RS at 2,5 and 7). Although pectin content in banana peel raw material decreased during ripening, pectin extraction was favoured. The highest alcohol-insoluble solids (AIS) yield (12.5%) was achieved at 70 °C, 15 mins from RS 7 peel. All extracts were homogalacturan-rich with some rhamnogalacturonan-I content (showing HGA/RG-I ratio > 2) with varied degree of methylation (DM). The highest HGA content (837.2 mg/g AIS) and HGA/RG-I ratio (9.9) were achieved at 110 °C, 0 mins from RS 7, suggesting its promising application as gelling agent. The highest RG-I content (111.1 mg/g AIS) were obtained at 110 °C, 5 mins from RS 7, which was comparable with the pectin with reported prebiotic ability isolated from the literature, suggesting its potential application in novel products.

Application of ionic liquid ultrasound-assisted extraction (IL-UAE) of lycopene from guava (Psidium guajava L.) by response surface methodology and artificial neural network-genetic algorithm

Lycopene-rich guava (Psidium guajava L.) exhibits significant economic potential as a functional food ingredient, making it highly valuable for the pharmaceutical and agro-food industries. However, there is a need to enhance the extraction methods of lycopene to fully exploit its beneficial uses. In this study, we evaluated various ionic liquids to identify the most effective one for extracting lycopene from guava. Among thirteen ionic liquids with varying carbon chains or anions, 1-butyl-3-methylimidazolium chloride demonstrated the highest productivity. Subsequently, a single-factor experiment was employed to test the impact of several parameters on the efficiency of lycopene extraction using this selected ionic liquid. These parameters included extraction time, ultrasonic power, liquid-solid ratio, concentration of the ionic liquid, as well as material particle size. Moreover, models of artificial neural networks using genetic algorithms (ANN-GA) and response surface methodology (RSM) were employed to comprehensively assess the first four key parameters. The optimized conditions for ionic liquid ultrasound-assisted extraction (IL-UAE) were determined as follows: 33 min of extraction time, 225 W of ultrasonic power, 22 mL/g of liquid-solid ratio, 3.0 mol/L of IL concentration, and extraction cycles of three. Under these conditions, lycopene production reached an impressive yield of 9.35 ± 0.36 mg/g while offering advantages such as high efficiency, time savings, preservation benefits, and most importantly environmental friendliness.

Optimization of soluble phosphate and IAA production using response surface methodology and ANN approach

Phosphorus (P) is often found inaccessible to plants, as it forms precipitates with cations and can be converted to accessible forms by using Phosphate solubilizing bacteria (PSB). In the present study, isolation and characterization of PSB from rhizospheric soil of coffee plants were performed. The influence of four independent variables (incubation temperature, incubation time, pH, and inoculum size) was investigated and optimized using an artificial neural network and response surface methodology on the solubility of phosphate and indole acetic acid production. The bacterium that can dissolve phosphate were isolated in Pikovskaya's agar containing insoluble tricalcium phosphate. Total, six Phosphate Solubilizing Bacteria were isolated and three of them (PSB1, PSB3, and PSB4) were found to be effectively solubilizing phosphate. Based on phosphate solubilizing index results Pseudomonas bacteria (PSB1) was selected for modeling. The results showed that both models performed reasonably well, but properly trained artificial neural networks have the more powerful modeling capability compared to the response surface method. The optimum conditions were found to be incubation temperature of 37.5 °C, incubation time of 9 days, pH of 7.2, and inoculum size of 1.89 OD. Under these conditions, the model predicted solubility of phosphate of 260.69 μg/ml and production of IAA of 80.00 μg/ml with a desirability value of 0.947. In general, the isolated Pseudomonas is expected to have phosphorus-degrading ability that promotes plant growth, and further field experimental work is required to use this bacterial strain as biofertilizer, as an alternative to synthetic fertilizer.

Optimization of substrate composition for pectinase production from Satkara (Citrus macroptera) peel using Aspergillus niger-ATCC 1640 in solid-state fermentation

Pectinase is an enzyme having a broad industrial and commercial application. However, higher production costs may be the major constraint for the wide-scale application of pectinase. Therefore, researchers are trying to reduce the pectinase production cost for subsequent application in the industrial processes by using a unique substrate and optimizing the fermentation medium components and process conditions. The main purpose of the current study was to optimize medium composition for pectinase production using Aspergillus niger-ATCC 1640 in the solid-state fermentation. The Response Surface Methodology (RSM) was performed to evaluate the effects of variables, specifically the concentrations of Satkara peel, urea, (NH₄)₂PO₄, NH₄NO₃, KH₂PO₄, ZnSO₄, and MgSO₄.7H₂O on pectinase production in the solid substrate. Firstly, a two-factorial design, Plackett-Burman design (PBD) was applied to screen the variables that significantly influenced the pectinase production. After finding the critical variables, 15 experimental runs were carried out using a Box-Behnken design (BBD) to derive a statistical model for optimizing the concentrations of the selected variables. The PBD model revealed that Satkara peel, urea, and (NH₄)₂SO₄ significantly affected the pectinase production. RSM results indicated that the predicted response for pectinase production was in good agreement with experimental data (R² = 0.9836). Under the optimized condition of Satkara peel (8.4 g/L), urea (0.5 g/L), and (NH₄)₂SO4 (2.7 g/L), the pectinase activity was predicted to be 0.6178 μmol/mL. In the present study, the experimental pectinase production achieved 0.6045 μmol/mL. The study revealed that optimization through RSM could improve the pectinase production from Satkara peel.

Modeling the influence of extraction parameters on the yield and chemical characteristics of microwave extracted mango (Mangifera indica L.) peel pectin by response surface methodology

Extraction is considered to be a critical unit operation to recover bioactive compounds from the in-situ state of many plant-based food processing wastes. The characteristics of pectin were predicted to vary with the source of raw material, extraction and post-extraction conditions. The study was focused to investigate the optimal conditions for extracting mango peel pectin (MPP) with increased yield and quality by microwave-assisted extraction (MAE). Box Behnken experimental design was used to model the influence of extraction parameters (microwave power, pH, and time) on the responses (yield, esterification degree, equivalent weight, anhydrouronic acid, and methoxyl content of pectin). The predicted models were adequately fitted to the experimental data (p ≤ 0.001) for all the response variables. A higher yield of pectin with better quality was obtained with optimal conditions of microwave power 606 watts (W), extraction time 5.15 min, and pH 1.83. The MPP obtained is categorized as low-methoxyl pectin as the range for the degree of esterification (DE) at all possible treatment variations remained below 50%. The study revealed that mango peel was an effective alternative source of pectin which could be extracted by microwave method on large scale.