Modeling the activity coefficient at infinite dilution of water in ionic liquids using artificial neural networks and support vector machines

Evaluation of AquaCrop and intelligent models in predicting yield and biomass values of wheat

AquaCrop is one of the dynamic and user-friendly models for simulating different conditions governing plant growth in the field. But this model requires many input parameters such as plant information, soil, climate, groundwater, and management factors. In this research, to solve this problem and develop a model with fewer input data, artificial neural network (ANN), support vector regression (SVR), and combined support vector regression with firefly algorithm (SVR-FFA) were used. For this purpose, 440 scenarios were created in 2 farms located in Iran, and the values of yield and biomass obtained by the AquaCrop model were compared with intelligent models. Also, consumable seed and irrigation were considered as inputs of intelligent models. The 99WestW2 farm in Miandoab had a seed yield of 6.588 t/ha, and the WestW10 farm in Mahabad had a seed yield of 5.05 t/ha. The results of this research showed that for both 99WestW2 and WestW10 farms, the SVR-FFA3 model was able to have the lowest amount of error so that for the amount of grain yield, the error values ​​for the farms were 0.033 and 0.069 t/ha, respectively. The error value of biomass was obtained for farms as 0.057 and 0.066 t/ha respectively. After SVR-FFA model, SVR and ANN models also showed good performance due to proper accuracy and saving time. Finally, SVR-FFA, SVR, and ANN models were able to predict yield and biomass values in the shortest time and with the highest accuracy with only two inputs.

Aleppo pine seeds (Pinus halepensis Mill.) as a promising novel green coagulant for the removal of Congo red dye: Optimization via machine learning algorithm

Consideration is now being given to the use of metal coagulants to remove turbidity from drinking water and wastewater. Concerns about the long-term impact of non-biodegradable sludge on human health and the potential contamination of aquatic systems are gaining popularity. Recently, alternative biocoagulants have been suggested to address these concerns. In this study, using a 1 M sodium chloride (NaCl) solution, the active coagulating agent was extracted from Pinus halepensis Mill. Seed, and used for the first time to remove Congo red dye, the influence of numerous factors on dye removal was evaluated in order to make comparisons with conventional coagulants. The application of biocoagulant was shown to be very successful, with coagulant dosages ranging from 3 to 12 mL L^-1 achieving up to 80% dye removal and yielding 28 mL L^-1 of sludge. It was also found that biocoagulant is extremely pH sensitive with an optimum operating pH of 3. Ferric chloride, on the other hand, achieved similar removal rate with higher sludge production (46 mL L^-1) under the same conditions. A Fourier Transform Infrared Spectroscopy and proximate composition analysis were undertaken to determine qualitatively the potential active coagulant ingredient in the seeds and suggested the involvement of proteins in the coagulation-flocculation mechanism. The evaluation criteria of the Support vector machine_Gray wolf optimizer model in terms of statistical coefficients and errors reveals quite interesting results and demonstrates the performance of the model, with statistical coefficients close to 1 (R = 0.9998, R² = 0.9995 and R² adj = 0.9995) and minimal statistical errors (RMSE = 0.5813, MSE = 0.3379, EPM = 0 0.9808, ESP = 0.9677 and MAE = 0.2382). The study findings demonstrate that Pinus halepensis Mill. Seed extract might be a novel, environmentally friendly, and easily available coagulant for water and wastewater treatment.

QSPR modeling of selectivity at infinite dilution of ionic liquids

The intelligent choice of extractants and entrainers can improve current mixture separation techniques allowing better efficiency and sustainability of chemical processes that are both used in industry and laboratory practice. The most promising approach is a straightforward comparison of selectivity at infinite dilution between potential candidates. However, selectivity at infinite dilution values are rarely available for most compounds so a theoretical estimation is highly desired. In this study, we suggest a Quantitative Structure–Property Relationship (QSPR) approach to the modelling of the selectivity at infinite dilution of ionic liquids. Additionally, auxiliary models were developed to overcome the potential bias from big activity coefficient at infinite dilution from the solute. Data from SelinfDB database was used as training and internal validation sets in QSPR model development. External validation was done with the data from literature. The selection of the best models was done using decision functions that aim to diminish bias in prediction of the data points associated with the underrepresented ionic liquids or extreme temperatures. The best models were used for the virtual screening for potential azeotrope breakers of aniline + n-dodecane mixture. The subject of screening was a combinatorial library of ionic liquids, created based on the previously unused combinations of cations and anions from SelinfDB and the test set extractants. Both selectivity at infinite dilution and auxiliary models show good performance in the validation. Our models’ predictions were compared to the ones of the COSMO-RS, where applicable, displaying smaller prediction error. The best ionic liquid to extract aniline from n-dodecane was suggested.