Modelling and optimization of crude oil removal from surface water via organic acid functionalized biomass using machine learning approach

Banana peel fiber adsorbent (BPF) with well-arranged substructure of pores was fabricated via esterification reaction with organic acid and biomass. The emerged adsorbent (BPF) was employed in taking away crude oil from water surface. Three machine learning tools such as RSM, ANN and ANFIS was employed for the modelling and optimization of the process. From results, the optimal oil layer removal of 98.2% was achieved at oil water ratio of 0.2 g /100 cm³. For now, BPF displayed high adsorptive prospect at a very low pH of 4 with 96.8% oil removal. On the other hand, the activation energy, enthalpy change and entropy change of the system are (18.56, 25.44, ?0.751 KJ/mols) and (25.77, 29.16, ?0.813 KJ/mols) designating a non-spontaneous system. The process of removal by BPF really matched the Langmuir isotherm model as proved by statistical error analysis with highest adsorption capacity of 49.33 mg/g as shown through equilibrium modeling. RSM displayed the optimum conditions of the key variables such as temperature, oil concentration, adsorbent dosage, pH and time as 100 °C, 0.2 g/100 cm³, 1.5 g, 2 and 75 mins, respectively. Analysis of the three generic algorithm indicated significant oil removal prediction with quite remarkably similar coefficient of correlation of 0.999. Additional statistical analysis suggested that RSM was marginally better than ANN and ANFIS for the modelling of crude oil removal via esterified banana peels fiber.     

Mixed phase Fe2O3/Mn3O4 magnetic nanocomposite for enhanced adsorption of methyl orange dye: Neural network modeling and response surface methodology optimization

In this research, a novel magnetic mesoporous adsorbent with mixed phase of Fe₂O₃/Mn₃O₄ nanocomposite was prepared by a facile precipitating method and characterized extensively. The prepared nanocomposite was used as adsorbent for toxic methyl orange (MO) dye removal from aqua matrix considering its high surface area (178.27 m²/g) with high saturation magnetization (23.07 emu/g). Maximum dye adsorption occurs at solution pH 2.0 and the electrostatic attraction between anionic form of MO dye molecules and the positively charged nanocomposite surface is the main driving force behind this adsorption. Response surface methodology (RSM) was used for optimizing the process variables and maximum MO removal of 97.67% is obtained at optimum experimental condition with contact time, adsorbent dose and initial MO dye concentration of 45 min, 0.87 g/l and 116 mg/l, respectively. Artificial neural network (ANN) model with optimum topology of 3–5–1 was developed for predicting the MO removal (%), which has shown higher predictive ability than RSM model. Maximum adsorption capacity of this nanocomposite was found to be 322.58 mg/g from Langmuir isotherm model. Kinetic studies reveal the applicability of second‐order kinetic model with contribution of intra‐particle diffusion in this process.     

Isotherm modelling and optimization of oil layer removal from surface water by organic acid activated plantain peels fiber

This research aimed to optimize and model the adsorption process of oil layer removal using activated plantain peels fiber (PPF), a biomass-based material. The adsorbent was activated by thermal and esterification methods using human and environmentally friendly organic acid. Effects of process parameters were examined by one factor at a time (OFAT) batch adsorption studies, revealing optimal conditions for oil removal. Also, RSM, ANN and ANFIS were used to adequately predict the oil removal with correlation coefficient > 0.98. RSM modelling revealed the best conditions as 90 °C, 0.2 mg/l, 1.5 g, 6 and 75 mins, for temperature, oil–water ratio, adsorbent dosage, pH and contact time respectively. Under these simulated conditions, the predicted oil removal was 96.88 %, which was experimentally validated as 97.44 %. Thermodynamic studies revealed the activation energy, change in enthalpy and change in entropy for irreversible pseudo-first order and pseudo-second order model as (15.82, 24.17, ?0.614 KJ/mols) and (33.21,40.31, ?0.106 KJ/mols) respectively, indicating non-spontaneous process; while modeling studies revealed that the adsorption process was highly matched to Langmuir’s isotherm, with maximum adsorption capacity of 50.34 mg/g. At the end of the overall statistical modelling, ANFIS performed marginally better than the ANN and RSM. It can be concluded from these results that our biomass-based material is an efficient, economically viable and sustainable adsorbent for oil removal, and has potentials for commercialization since the process of adsorption highly matched with standard models, and its capacity or percentage oil removal also compares favorably to that of commercially available adsorbents.     

Synthesis of carbon nanotube‐based nanocomposite and application for wastewater treatment by ultrasonicated adsorption process

The sorption of methylene blue (MB) and basic yellow 28 (BY28) dyes in water on Ag@ZnO/MWCNT (Ag‐doped ZnO loaded on multiwall carbon nanotubes) nanocomposite is investigated in a batch process, optimizing starting initial dye concentration, sonication time and adsorbent mass. Isotherms and kinetic behaviours of MB and BY28 adsorption onto Ag@ZnO/MWCNT were explained by extended Freundlich and pseudo‐second‐order kinetic models. Ag@ZnO/MWCNT was synthesized and characterized using X‐ray diffraction, energy‐dispersive X‐ray spectroscopy, field emission scanning electron microscopy and Brunauer–Emmett–Teller analysis. According to the experimental data, adaptive neuro‐fuzzy inference system (ANFIS), generalized regression neural network (GRNN), backpropagation neural network (BPNN), radial basic function neural network (RBFNN) and response surface methodology (RSM) were developed, and applied to forecast the removal performance of the sorbent. The influence of process variables (i.e. sonication time, initial dye concentration, adsorbent mass) on the removal of MB and BY28 was considered by central composite rotatable design of RSM, GRNN, ANFIS, BPNN and RBFNN. The performances of the developed ANFIS, GRNN, BPNN and RBFNN models were compared with RSM mathematical models in terms of the root mean square error, coefficient of determination, absolute average deviation and mean absolute error. The coefficients of determination calculated from the validation data for ANFIS, GRNN, BPNN, RBFNN and RSM models were 0.9999, 0.9997, 0.9883, 0.9898 and 0.9608 for MB and 0.9997, 0.9990, 0.9859, 0.9895 and 0.9593 for BY28 dye, respectively. The ANFIS model was found to be more precise compared to the other models. However, the GRNN method is much easier than the ANFIS method and needs less time for analysis. So, it has potential in chemometrics and it is feasible that the GRNN algorithm could be applied to model real systems. The monolayer adsorption capacity of MB and BY28 was 292.20 and 287.02 mg g^?1, respectively.     

Application of artificial neural network for comparison and modeling of the ultrasonic and stirrer assisted removal of anionic dye using activated carbon supported with nanostructure material

In this study, a green approach has been described for the synthesis of copper sulfide nanoparticles loaded on activated carbon (CuS‐NP‐AC) and usability of it for the removal of sunset yellow (SY) dye by ultrasound‐assisted and stirrer has been compared. In addition, the artificial neural network (ANN) model has been employed for a forecasting removal percentage of SY dye using the results obtained. This material was characterized using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The impact of variables, including initial dye concentration (mg/L), pH, adsorbent dosage (g), sonication time (min) and temperature (°C) on SY removal was studied. Fitting the experimental equilibrium data of different isotherm models such as Langmuir, Freundlich, Temkin and Dubinin–Radushkevich models display the suitability and applicability of the Langmuir model. Analysis of experimental adsorption data of different kinetic models including pseudo‐first and second order, Elovich and intraparticle diffusion models indicate the applicability of the second‐order equation model. The adsorbent (0.005 g) is applicable for successful removal of SY dye (> 98%) in short time (9 min) under ultrasound condition. A three layer ANN models with 8 and 6 neurons at hidden layer was selected as optimal models using stirrer and ultrasonic, respectively. These models displayed a good agreement between forecasted data and experimental data with the determination coefficient (R²) of 0.9948 and 0.9907 and mean squared error (MSE) of 0.0001 and 0.0002 for training set using stirrer and ultrasonic, respectively.     

Enhanced adsorption performance of a novel Fe‐Mn‐Zr metal oxide nanocomposite adsorbent for anionic dyes from binary dye mix: Response surface optimization and neural network modeling

A novel adsorbent, Fe‐Mn‐Zr metal oxide nanocomposite was synthesized and investigated for removal of methyl orange (MO) and eosin yellow (EY) dyes from binary dye solution. The magnetic nanocomposite has shown surface area of 143.01 m²/g and saturation magnetization of 15.29 emu/g. Optimization was carried out via response surface methodology (RSM) for optimizing process variables, and optimum dye removal of 99.26% and 99.55% were obtained for MO and EY dye, respectively with contact time 62 min, adsorbent dose 0.45 g/l, initial MO concentration 11.0 mg/l, and initial EY concentration 25.0 mg/l. A feed forward back propagation neural network model has shown better prediction ability than RSM model for predicting MO and EY dye removal (%). Adsorption process strictly follows Langmuir isotherm model, and enhanced adsorption capacities of 196.07 and 175.43 mg/g were observed for MO and EY dye, respectively due to synergistic effects of physicochemical properties of trimetal oxides. Surface adsorption and pore diffusions are the mechanisms involved in the adsorption as revealed from kinetic studies.     

Multi‐responses optimization of simultaneous adsorption of methylene blue and malachite green dyes in binary aqueous system onto Ni:FeO(OH)‐NWs‐AC using experimental design: derivative spectrophotometry method

In this research, response surface methodology (RSM) approach using Central Composite Design (CCD) coupled by derivative spectrophotometry method was applied to develop mathematical model and optimize process parameters for simultaneous adsorption of methylene blue (MB) and malachite green (MG) from aqueous solution using Ni:FeO(OH) ‐ NWs‐AC. The optimal conditions to adsorption of MB and MG in binary mixture solution from aqueous solution were found at pH 8.0, MB concentration 20 mg L^‐1, MG concentration 20 mg L^‐1, adsorbent dosage 0.033 g and contact time 40 min. At these conditions, high adsorption efficiency (99.39% and 100.0% for MB and MG, respectively) was achieved. Among experimental equilibrium, Langmuir isotherm model fitted well with maximum monolayer adsorption capacity of 28.6 and 29.8 mg g^‐1 for MB and MG, respectively. The adsorption kinetic data followed pseudo second‐order kinetics for MB and MG dyes.     

Gold Nanoparticles Loaded on Activated Carbon as Novel Adsorbent for Kinetic and Isotherm Studies of Methyl Orange and Sunset Yellow Adsorption

In this research, a novel adsorbent gold nanoparticle loaded on activated carbon (Au-NP-AC) was synthesized by a low cost in a routine protocol. Subsequently, this novel material characterization and identification are followed by different techniques such as th eBruner–Emmet–Teller (BET) theory, scanning electron microcopy, and transmission electron microscopy analysis. Unique properties such as high BET surface area (>1229.55 m²/g) and low pore size (<22.46 Å) and average particle size lower than 48.798 Å in addition to high reactive atom and presence of various functional groups make it possible for efficient removal of sunset yellow (SY) and methyl orange (MO). Generally, the influence of variables including amount of adsorbent, initial dyes concentration, contact time, temperature on dyes removal percentage has great effect on removal percentage that their influence was optimized. The kinetic of proposed adsorption processes efficiently followed, pseudo-second-order and intra-particle diffusion approach. The equilibrium data of the removal strongly follow the Langmuir monolayer adsorption with high adsorption capacity in a short amount of time. This novel adsorbent by small amount (0.01 g) really is applicable for removal of high amount of both dyes (MO and SY) in short time (<18 minutes). Equilibrium data fitted well with the Langmuir model at all amount of adsorbent, while maximum adsorption capacity for MO 161.29 mg g^?1 and for SY 227.27 for 0.005 g of Au-NP-AC.     

An electronic tongue using potentiometric all-solid-state PVC-membrane sensors for the simultaneous quantification of ammonium and potassium ions in water

The simultaneous determination of NH₄⁺ and K⁺ in solution has been attempted using a potentiometric sensor array and multivariate calibration. The sensors used are rather non-specific and of all-solid-state type, employing polymeric (PVC) membranes. The subsequent data processing is based on the use of a multilayer artificial neural network (ANN). This approach is given the name "electronic tongue" because it mimics the sense of taste in animals. The sensors incorporate, as recognition elements, neutral carriers belonging to the family of the ionophoric antibiotics. In this work the ANN type is optimized by studying its topology, the training algorithm, and the transfer functions. Also, different pretreatments of the starting data are evaluated. The chosen ANN is formed by 8 input neurons, 20 neurons in the hidden layer and 2 neurons in the output layer. The transfer function selected for the hidden layer was sigmoidal and linear for the output layer. It is also recommended to scale the starting data before training. A correct fit for the test data set is obtained when it is trained with the Bayesian regularization algorithm. The viability for the determination of ammonium and potassium ions in synthetic samples was evaluated; cumulative prediction errors of approximately 1% (relative values) were obtained. These results were comparable with those obtained with a generalized regression ANN as a reference algorithm. In a final application, results close to the expected values were obtained for the two considered ions, with concentrations between 0 and 40 mmol L^–1.     

Optimization of artificial neural network for retention modeling in high-performance liquid chromatography

An artificial neural network (ANN) model for the prediction of retention times in high-performance liquid chromatography (HPLC) was developed and optimized. A three-layer feed-forward ANN has been used to model retention behavior of nine phenols as a function of mobile phase composition (methanol-acetic acid mobile phase). The number of hidden layer nodes, number of iteration steps and the number of experimental data points used for training set were optimized. By using a relatively small amount of experimental data (25 experimental data points in the training set), a very accurate prediction of the retention (percentage normalized differences between the predicted and the experimental data less than 0.6%) was obtained. It was shown that the prediction ability of ANN model linearly decreased with the reduction of number of experiments for the training data set. The results obtained demonstrate that ANN offers a straightforward way for retention modeling in isocratic HPLC separation of a complex mixture of compounds widely different in pK_a and log K_ow values.     

Design of optimum response surface experiments for the adsorption of acetic,butyric, and oxalic acids on Amberlyst A21

In this study the adsorption equilibria of acetic, butyric, and oxalic acids onto Amberlyst A21 were investigated experimentally at 25°C. The process was optimized using response surface methodology (RSM). The time to reach the equilibrium state, effects of adsorbent amount, and initial acid concentrations on adsorption efficiency were investigated. Freundlich, Langmuir, and Temkin isotherms were applied to experimental data. The Freundlich isotherm revealed better results than the others. In addition to the main aim of this research, a statistical/mathematical approach – RSM – was utilized to simulate and determine the optimum conditions of acetic, butyric, and oxalic acids removal by Amberlyst A21 using three selected parameters (adsorbent dose, initial dye concentration, and type of acids). The significance of independent variables and their interactions were tested by the analysis of variance (ANOVA). The optimum acid concentration, amount of adsorbent, type of acid, and removal of acid (%) were found by desirability function to be 0.199?mol/L, 1.999?g, butyric acid, and 84.537%, respectively.     

Synthesis of CuS nanoparticles loaded on activated carbon composite for ultrasound‐assisted adsorption removal of dye pollutants: Process optimization using CCD‐RSM,equilibrium and kinetic studies

In this study, the CuS nanoparticles loaded on activated carbon (CuS‐NPs‐AC) composite was synthesized and then, characterized by XRD and FE‐SEM analyses. The prepared composite was used as a potential adsorbent for the simultaneous ultrasound‐assisted removal of Indigo Carmine (IC) and Safranin‐O (SO). The CuS‐NPs‐AC dose (0.01‐0.03 g), sonication time (1‐5 min), initial SO concentration (5‐15 mg L^‐1) and initial IC concentration (5‐15 mg L^‐1) as expectable effective parameters were studied by central composite design (CCD) under response surface methodology (RSM) to obtain an useful knowledge about the effect of simultaneous interaction between IC and SO on their removal percentage. The optimum SO and IC removal percentages were determined to be 98.24 and 97.15% at pH = 6, 0.03 g of the CuS‐NPs‐AC, 3 min sonication time, 12 and 10 mg L^‐1 of IC and SO. The values of coefficient of determination (R²) for SO and IC were 0.9608 and 0.9796, respectively, indicating the favorable fitness of the experimental data to the second order polynomial regression model. The isotherm data were well correlated with Freundlich model. The maximum monolayer adsorption capacities of 87.5 and 69.90 mg g^‐1 at room temperature for IC and SO in the investigated binary system expressed the high efficiency of the novel adsorbent for water cleanup within a short time. The investigation of correlation between time and rate of adsorption revealed that IC and SO adsorption onto the CuS‐NPs‐AC followed pseudo‐second‐order and intra‐particle diffusion simultaneously.     

Experimental design,machine learning approaches for the optimization and modeling of caffeine adsorption

In the current research, the sorption of caffeine on fresh and calcined Cu–Al layered double hydroxide was comparatively studied based on adsorption parameters, adsorption kinetics, and adsorption isotherm. Response surface methodology (RSM), support vector machine (SVM) and artificial neural network (ANN), as data mining methods, were applied to develop models by considering various operating variables. Different characterization methods were exploited to conduct a comprehensive analysis of the characteristics of HDL in order to acquire a thorough understanding of its structural and functional features. The Langmuir model was employed to accurately describe the maximum monolayer adsorption capacity for calcined sample (q_max) of 152.99 mg/g mg/g with R² = 0.9977. The pseudo-second order model precisely described the adsorption phenomenon (R² = 0.999). The thermodynamic analysis also reveals a favorable and spontaneous process. The ANN model predicts adsorption efficiency result with R² = 0.989. The five-fold cross-validation was achieved to evaluate the validity of the SVM. The predication results revealed approximately 99.9% accuracy for test datasets and 99.63% accuracy for experiment data. Moreover, ANOVA analysis employing the central composite design-response surface methodology (CCD-RSM) indicated a good agreement between the quadratic equation predictions and the experimental data, which results in R² of 0.9868 and the highest removal percentages in optimized step were obtained for RSM (pH 5.05, mass of adsorbent 20 mg, time of 72 min, and caffeine concentrations of 22 mg/L). On the whole, the findings confirm that the proposed machine learning models provided reliable and robust computer methods for monitoring and simulating the adsorption of pollutants from aqueous solutions by Cu–Al–LDH.     

Synthesis and characterization of SnO2/(NH4)2‐SnCl6 nanocomposites loaded on activated carbon and its application for adsorption of methylene Blue and Orange G

The present study deals with the synthesis and characterization (FE‐SEM, particle size distribution, XRD and point of zero charge) SnO₂/(NH₄)₂‐SnCl₆ nanocomposites loaded on activated carbon (SnO₂/(NH₄)₂‐SnCl₆‐NCs‐AC) and its subsequent application for the simultaneous removal of Methylene Blue (MB) and Orange G (OG) from aqueous solution. Response surface methodology (RSM) based on central composite design (CCD) give trend of influencing responses with respect to five parameters such as contact time (X₁), OG concentration (X₂), MB concentration (X₃), adsorbent mass (X₄) and pH (X₅). In later stage following recognition of significant variables and interaction, quadratic model generated which are able to predict the dyes removal in different conditions. Justification and selection of significant terms was conducted based on analysis of variance and Fisher's F‐test Optimal value of contact time, OG concentration, MB concentration, adsorbent mass and pH were set at 4.0 min, 10 mg l^?1, 20 mg L^?1, 0.015 g and 6.0, respectively, which lead to achievement of best experiment removal percentage of 97.0 and 99.5% OG and MB respectively, from their binary solutions. The whole experimental data follow pseudo‐first‐order and pseudo‐second‐order rate equations. The fitting experimental data to more available conventional model like Langmuir, Freundlich, Temkin and Dubinin‐Radushkevich isotherm models revel more ability of Langmuir model (with R² > 0.997) for explanation of system in equilibrium. The adsorption efficiency remained high even after the five cycle of reuse (99.76% and 95.56% for MB and OG, respectively).     

Homogeneous liquid-liquid microextraction via flotation assistance coupled with gas chromatography-mass spectrometry for determination of myclobutanil in cucumber,tomato, grape,and strawberry using genetic algorithm

Facile and potent homogeneous liquid–liquid microextraction via flotation assistance method (HLLME-FA) combined with gas chromatography-mass spectrometry was proposed for determination of trace amounts of myclobutanil in fruit and vegetable samples. The paramount parameters, such as extraction and homogeneous solvent types and volumes, ionic strength and extraction time were studied. Under optimum conditions, the detection limit of 0.005 ng g^?1, the linear range of 0.05–100 ng g^?1, and the precision of 3.8% were acquired. A three-layer arti?cial neural network (ANN) model was used with 10 neurons and tan-sigmoid function at hidden layer and a linear transfer function at output layer were developed to predict the process. The results indicated that the proposed ANN model could perfectly predict the process with the mean square error of 0.89%. Then genetic algorithm was utilised to optimise the parameters. The proposed procedure showed satisfactory results for analysis of cucumber, tomato, grape, and strawberry.     

Thermodynamic study of Iron (III) removing by the synthesized α-Alumina powder and evaluating the corresponding adsorption isotherm models using Response Surface Method

In this study, α-Alumina as an adsorbent was initially synthesized by combustion method and characterized by scanning electron microscope (SEM), transmission electron microscopy (TEM), X-ray diffraction (XRD), EDAX and Brunauer–Emmett–Teller (BET) techniques. Then the efficiency of the synthetized adsorbent for iron (III) removal was investigated and the effect of corresponding parameters such as adsorbent dose, contact time, initial concentration, temperature and solution pH on the adsorption capacity were examined and the optimums values of these parameters were concluded upon the surfaces response method. By using response surface methodology (RSM), the adsorption experimental design was performed and the statistical analysis showed that the quadratic model as well as the model terms were significant. In addition, the experimental results were examined with some suitable models, such as Langmuir, Freundlich isotherm models, where Freundlich model fitted better our experimental results. Finally, the thermodynamic behavior of the studied adsorption process was considered and the thermodynamic functions of the process were evaluated. The results showed that the Fe (III) ion adsorption onto the synthesized adsorbent is exothermic and spontaneous at the experimental conditions.     

Application of artificial neural networks for formulation and modeling of dye adsorption onto multiwalled carbon nanotubes

In this study, an artificial neural network (ANN) has been developed to predict the adsorption amount of dye (methylene blue) onto multiwalled carbon nanotubes. Batch experiments have been carried out to obtain experimental data. Important parameters in the adsorption system such as initial dye concentration, adsorbent dosage, temperature, pH and contact time have been used as the inputs of the network, while the output is the final concentration of dye in aqueous solution after adsorption. The neural network structure has been optimized by testing various training algorithms and different number of neurons in a hidden layer. An empirical equation for determination of final dye concentration in aqueous solutions after adsorption has been developed by using the weights of the optimized network. The results of the optimized ANN have been compared with conventional models in equilibrium and kinetic fields. According to error analysis and determination coefficient, the ANN was found to be the most appropriate model to describe this adsorption process. Sensitivity analysis showed that initial dye concentration, pH and contact time are the most effective parameters in this process. The influence percentages of these parameters on the output were 28, 24 and 24 %, respectively.