Prediction of arsenic (III) adsorption from aqueous solution using non-neural network algorithms

Abstract

Heavy metals such as arsenic can be effectively removed through adsorption. Through material property evaluation and adsorption parameter optimization, machine learning (ML) modeling provides an alternative to lengthy laboratory experimentation. In this work, adsorption data from an earlier study employing a waste-material composite were used. To create prediction models, four non-neural network algorithms - support vector machines (SVM), Gaussian process regression (GPR), linear regression, and ensemble approaches - were used and contrasted with neural network algorithms. Nine predictors were utilized, ranging from adsorbent composition alterations to experimental circumstances. Using principal component analysis (PCA) and feature selection, together with the F-test and minimum redundancy maximum relevance (MRMR) algorithms for feature reduction, optimization was accomplished. With an R-squared of 0.939, mean absolute error (MAE) of 5.778, and root mean squared error (RMSE) of 7.119 for training and an R-squared of 0.942, MAE of 5.450, and RMSE of 6.870 for testing, the optimized GPR method offered the best predictive performance. The best R-squared values found for other algorithms were: SVM (0.922), linear regression (0.925), and ensemble (0.927). The most important variables influencing adsorption efficiency were initial arsenic concentration, time, and the iron salt content. Local interpretable model-agnostic explanations (LIME), partial dependence plot (PDP), and Shapley additive explanations (SHAP) plots were used to explain these results. This work shows that, based on model-derived parameters, non-neural network algorithms may efficiently simulate and optimize arsenic adsorption tests, providing a trustworthy substitute for neural network techniques and markedly increasing adsorption efficiency.

Keywords

Adsorption / arsenic removal / machine learning / non-neural network regression / water quality

References

[1] Ren X,Nagatsu M.Carbon nanotubes as adsorbents in environmental pollution management: a review.Chem Eng J2011;170:395-410

[2] Chung JY,Hong YS.Environmental source of arsenic exposure.J Prev Med Public Health2014;47:253-7 PMCID:PMC4186553

[3] Ravenscroft P,Richards K.Arsenic pollution: a global synthesis. Oxford, UK: Wiley-Blackwell, 2009.

[4] Smedley P.A review of the source, behaviour and distribution of arsenic in natural waters.Appl Geochem2002;17:517-68

[5] Chutia P,Kojima T.Arsenic adsorption from aqueous solution on synthetic zeolites.J Hazard Mater2009;162:440-7

[6] Song W,Liang J.Removal of As(V) from wastewater by chemically modified biomass.J Mol Liq2015;206:262-7

[7] Sigdel A,Kwak H.Arsenic removal from aqueous solutions by adsorption onto hydrous iron oxide-impregnated alginate beads.J Ind Eng Chem2016;35:277-86

[8] Sun J,Zhang A.Preparation of Fe-Co based MOF-74 and its effective adsorption of arsenic from aqueous solution.J Environ Sci2019;80:197-207

[9] Wang C,Wu C.Metal-organic frameworks for aquatic arsenic removal.Water Res2019;158:370-82

[10] Holm TR.Effects of CO₃^2-/bicarbonate, Si, and PO₄^3- on Arsenic sorption to HFO.J AWWA2002;94:174-81

[11] Bissen M.Arsenic - a review. Part II: oxidation of Arsenic and its removal in water treatment.Acta hydrochim hydrobiol2003;31:97-107

[12] Choong TS,Robiah Y,Azni I.Arsenic toxicity, health hazards and removal techniques from water: an overview.Desalination2007;217:139-66

[13] Norberto J,Mcphedran KN.Microwave activated and iron engineered biochar for arsenic adsorption: life cycle assessment and cost analysis.J Environ Chem Eng2023;11:109904

[14] Masue Y,Kramer TA.Arsenate and arsenite adsorption and desorption behavior on coprecipitated aluminum:iron hydroxides.Environ Sci Technol2007;41:837-42

[15] Mohan D.Arsenic removal from water/wastewater using adsorbents - a critical review.J Hazard Mater2007;142:1-53

[16] Lee B,Lee H.Synthesis of functionalized porous silicas via templating method as heavy metal ion adsorbents: the introduction of surface hydrophilicity onto the surface of adsorbents.Micropor Mesopor Mat2001;50:77-90

[17] Ahmaruzzaman M.Industrial wastes as low-cost potential adsorbents for the treatment of wastewater laden with heavy metals.Adv Colloid Interface Sci2011;166:36-59

[18] Zhang L,Cheng Z.Removal of heavy metal ions using chitosan and modified chitosan: a review.J Mol Liq2016;214:175-91

[19] Tizhe B,Nkafamiya I.Biosorption of metal ions from aqueous solution by immobilized moringa oleifera bark.IRJPAC2015;5:238-44

[20] Barua S,Nazimuddin M.Evaluation of Moringa oleifera carbon for the As(III) removal from contaminated groundwater.Int J Innov Appl Stud2014;8:1390-9Available from: https://www.cabidigitallibrary.org/doi/full/10.5555/20153089415. [Last accessed on 7 Dec 2024]

[21] Mirza NH.Gypsum-based porous media filter with solid waste: a novel composite for removing arsenic (III) from aqueous solution.Int J Eng2025;38:735-43

[22] Ouyang D,Hu L,Hu Y.Research on the adsorption behavior of heavy metal ions by porous material prepared with silicate tailings.Minerals2019;9:291

[23] Mazloom G,Khorasheh F.Kinetic modeling of pyrolysis of scrap tires.J Anal Appl Pyrol2009;84:157-64

[24] Khraibet SA,Banisharif F.Comparative study of different two-phase models for the propane oxidative dehydrogenation in a bubbling fluidized bed containing the VO_x/γ-Al₂O₃ catalyst.Ind Eng Chem Res2021;60:9729-38

[25] Jaffari ZH,Kim CM.Transformer-based deep learning models for adsorption capacity prediction of heavy metal ions toward biochar-based adsorbents.J Hazard Mater2024;462:132773

[26] Yaseen ZM.An insight into machine learning models era in simulating soil, water bodies and adsorption heavy metals: Review, challenges and solutions.Chemosphere2021;277:130126

[27] Xiong T,Hou Z.Prediction of arsenic adsorption onto metal organic frameworks and adsorption mechanisms interpretation by machine learning.J Environ Manage2023;347:119065

[28] Liu R,Zhang P,Tao D.A deep learning neural network approach for predicting the factors influencing heavy-metal adsorption by clay minerals.Clay Miner2022;57:70-6

[29] Shaheen SM,Hassan NE.Wood-based biochar for the removal of potentially toxic elements in water and wastewater: a critical review.Int Mater Rev2019;64:216-47

[30] Mongioví C,Gabrion X.Revealing the adsorption mechanism of copper on hemp-based materials through EDX, nano-CT, XPS, FTIR, Raman, and XANES characterization techniques.Chem Eng J Adv2022;10:100282

[31] Mallik AK,Rahman MA,Rahman MM.Progress in surface-modified silicas for Cr(VI) adsorption: a review.J Hazard Mater2022;423:127041

[32] Ismail UM,Vohra MS.Aqueous Pb(II) removal using ZIF-60: adsorption studies, response surface methodology and machine learning predictions.Nanomaterials2023;13:1402 PMCID:PMC10141474

[33] Abdi J.Machine learning approaches for predicting arsenic adsorption from water using porous metal-organic frameworks.Sci Rep2022;12:16458 PMCID:PMC9525301

[34] Aftab RA,Danish M,Danish M.Novel machine learning (ML) models for predicting the performance of multi-metal binding green adsorbent for the removal of Cd (II), Cu (II), Pb (II) and Zn (II) ions.Environ Adv2022;9:100256

[35] Zhu X,Tsang DC.Machine learning for the selection of carbon-based materials for tetracycline and sulfamethoxazole adsorption.Chem Eng J2021;406:126782

[36] Yin G,Mironov S.Machine learning method for simulation of adsorption separation: comparisons of model’s performance in predicting equilibrium concentrations.Arab J Chem2022;15:103612

[37] Dashti A,Riasat Harami H,Asghari M.Biochar performance evaluation for heavy metals removal from industrial wastewater based on machine learning: application for environmental protection.Sep Purif Technol2023;312:123399

[38] Almalawi A,Alqurashi F,Alam MM.Modeling of remora optimization with deep learning enabled heavy metal sorption efficiency prediction onto biochar.Chemosphere2022;303:135065

[39] Elbana TA,Akrami N,Shaheen SM.Freundlich sorption parameters for cadmium, copper, nickel, lead, and zinc for different soils: influence of kinetics.Geoderma2018;324:80-8

[40] Ducamp M.Prediction of thermal properties of zeolites through machine learning.J Phys Chem C2022;126:1651-60

[41] Ma S.Machine learning for atomic simulation and activity prediction in heterogeneous catalysis: current status and future.ACS Catal2020;10:13213-26

[42] Mandal S,Sahu M.Artificial neural network modelling of As(III) removal from water by novel hybrid material.Process Saf Environ2015;93:249-64

[43] Sakizadeh M.Artificial intelligence for the prediction of water quality index in groundwater systems.Model Earth Syst Environ2016;2:63

[44] Guo Y,Shawe-Taylor J.Covering numbers for support vector machines.IEEE Trans Inform Theory2002;48:239-50

[45] Durbha SS,Younan NH.Support vector machines regression for retrieval of leaf area index from multiangle imaging spectroradiometer.Remote Sens Environ2007;107:348-61

[46] Čeh M,Lisec A.Estimating the performance of random forest versus multiple regression for predicting prices of the apartments.IJGI2018;7:168

[47] Wei L,Zhong Y,Hu X.An improved gradient boosting regression tree estimation model for soil heavy metal (arsenic) pollution monitoring using hyperspectral remote sensing.Appl Sci2019;9:1943

[48] Cha Y,Choi JW,Cho KH.Bayesian modeling approach for characterizing groundwater arsenic contamination in the Mekong River basin.Chemosphere2016;143:50-6

[49] Jang JSR.ANFIS: adaptive-network-based fuzzy inference system.IEEE Trans Syst Man Cybern1993;23:665-85

[50] Abdullahi J,Rimtip NN,Aslanova F.A novel approach for precipitation modeling using artificial intelligence-based ensemble models.Desalin Water Treat2024;317:100188

[51] Mirza NH.Aqueous arsenic (III) removal using a novel solid waste based porous filter media block: traditional and machine learning (ML) approaches.Desalin Water Treat2024;319:100536

[52] Zhu JJ,Ren ZJ.Machine learning in environmental research: common pitfalls and best practices.Environ Sci Technol2023;57:17671-89

[53] Zhao S,Chen C,Tao J.Interpretable machine learning for predicting and evaluating hydrogen production via supercritical water gasification of biomass.J Clean Prod2021;316:128244

[54] Li J,Tong YW.Understanding and optimizing the gasification of biomass waste with machine learning.Green Chem Eng2023;4:123-33

[55] Alemneh ST,Hitzmann B.Comparative study of chemical composition, pasting, thermal and functional properties of Teff (Eragrostis tef) flours grown in Ethiopia and South Africa.Int J Food Prop2022;25:144-58

[56] Smith A,Dumesic JA,Zavala VM.A machine learning framework for the analysis and prediction of catalytic activity from experimental data.Appl Catal B Environ2020;263:118257

[57] Li J,Li C.Data-driven based in-depth interpretation and inverse design of anaerobic digestion for CH₄-rich biogas production.ACS EST Eng2022;2:642-52

[58] Jamei M,Alawi OA.Earth skin temperature long-term prediction using novel extended Kalman filter integrated with artificial intelligence models and information gain feature selection.Sustain Comput Infor2022;35:100721

[59] Breiman L.Random forests.Mach Learn2001;45:5-32

[60] Zhu X,Ok YS.The application of machine learning methods for prediction of metal sorption onto biochars.J Hazard Mater2019;378:120727

[61] Pathy A,P B.Predicting algal biochar yield using eXtreme Gradient Boosting (XGB) algorithm of machine learning methods.Algal Res2020;50:102006

[62] Kooh MRR,Chou Chau Y,Lim CM.Machine learning approaches to predict adsorption capacity of Azolla pinnata in the removal of methylene blue.J Taiwan Inst Chem Eng2022;132:104134

[63] Ghaedi M,Ahmadi A.Artificial neural network-genetic algorithm based optimization for the adsorption of phenol red (PR) onto gold and titanium dioxide nanoparticles loaded on activated carbon.J Ind Eng Chem2015;21:587-98

[64] Zhang S,Liu C.Modeling and optimization of porous aerogel adsorbent for removal of cadmium from crab viscera homogenate using response surface method and artificial neural network.LWT2021;150:111990

[65] Jiang H,Li Z.Non-parallel hyperplanes ordinal regression machine.Knowl Based Syst2021;216:106593

[66] Emmert-Streib F.Taxonomy of machine learning paradigms: a data-centric perspective.WIREs Data Min Knowl2022;12:e1470

[67] Ye H,Yang T.Comparative study on the performance prediction of fuel cell using support vector machine with different kernel functions. In: Proceedings of China SAE Congress 2018: Selected Papers; Singapore. 2020. pp. 337-51.

[68] Lee SH,Wang X.Online-learning-aided optimization and interpretation of sugar production from oil palm mesocarp fibers with analytics for industrial applications.Resour Conserv Recy2022;180:106206

[69] Zhang L,Zhang X.Modeling and optimization of microbial lipid fermentation from cellulosic ethanol wastewater by Rhodotorula glutinis based on the support vector machine.Bioresour Technol2020;301:122781

[70] Da T,He W,Chen T.Prediction of uranium adsorption capacity on biochar by machine learning methods.J Environ Chem Eng2022;10:108449

[71] Li J,Liu T.Machine learning aided bio-oil production with high energy recovery and low nitrogen content from hydrothermal liquefaction of biomass with experiment verification.Chem Eng J2021;425:130649

[72] Li J,Pan L,Wang X.A hybrid data-driven and mechanistic modelling approach for hydrothermal gasification.Appl Energy2021;304:117674

[73] Cai J,Zhu Y,Li L.Prediction and analysis of net ecosystem carbon exchange based on gradient boosting regression and random forest.Appl Energy2020;262:114566

[74] Dehghanian N,Ansari A.A random forest approach for predicting the removal of Congo red from aqueous solutions by adsorption onto tin sulfide nanoparticles loaded on activated carbon.Desalin Water Treat2016;57:9272-85

[75] He Z,Rao P.WS₂ significantly enhances the degradation of sulfachloropyridazine by Fe(III)/persulfate.Sci Total Environ2022;850:157987

[76] Haq MA.DNNBoT: deep neural network-based botnet detection and classification.Comput Mater Con2022;71:1729-50

[77] Koeshidayatullah A.Optimizing image-based deep learning for energy geoscience via an effortless end-to-end approach.J Pet Sci Eng2022;215:110681

[78] Fan Y.A backpropagation learning algorithm with graph regularization for feedforward neural networks.Inform Sci2022;607:263-77

[79] Yaqoob U.Chemical gas sensors: recent developments, challenges, and the potential of machine learning - a review.Sensors2021;21:2877

[80] Yuan X,Lim JY.Machine learning for heavy metal removal from water: recent advances and challenges.ACS EST Water2024;4:820-36

[81] Alam MA,Alam MO.Adsorption of As (III) and As (V) from aqueous solution by modified Cassia fistula (golden shower) biochar.Appl Water Sci2018;8:839

[82] Nguyen TH,Nguyen Thi HT.Synthesis of iron-modified biochar derived from rice straw and its application to arsenic removal.J Chem2019;2019:1-8

[83] Yang S,Aierken A.Mono/competitive adsorption of Arsenic(III) and Nickel(II) using modified green tea waste.J Taiwan Inst Chem Eng2016;60:213-21

[84] Imran M,Iqbal J.Synthesis, characterization and application of novel MnO and CuO impregnated biochar composites to sequester arsenic (As) from water: modeling, thermodynamics and reusability.J Hazard Mater2021;401:123338

[85] Giles DE,Issa TB,Singh P.Iron and aluminium based adsorption strategies for removing arsenic from water.J Environ Manage2011;92:3011-22

[86] Sumathi T.Adsorption studies for arsenic removal using activated Moringa oleifera.Int J Chem Eng2014;2014:1-6

[87] Bae J,Kim KS,Choi H.Adsorptive removal of arsenic by mesoporous iron oxide in aquatic systems.Water2020;12:3147

[88] Mudzielwana R,Ndungu P.Removal of As(III) from synthetic groundwater using Fe-Mn bimetal modified kaolin clay: adsorption kinetics, isotherm and thermodynamics studies.Environ Process2019;6:1005-18

[89] Islam SR,Bundy S. Infusing domain knowledge in AI-based “black box” models for better explainability with application in bankruptcy prediction. arXiv. [Preprint.] May 30, 2019 [accessed 2024 Dec 7]. Available from: https://doi.org/10.48550/arXiv.1905.11474.

[90] Li J,Suvarna M,Wang X.Fuel properties of hydrochar and pyrochar: prediction and exploration with machine learning.Appl Energy2020;269:115166

[91] Goldstein A,Bleich J.Peeking inside the black box: visualizing statistical learning with plots of individual conditional expectation.J Comput Graphical Stat2015;24:44-65

[92] Yuan X,Low S.Applied machine learning for prediction of CO₂ adsorption on biomass waste-derived porous carbons.Environ Sci Technol2021;55:11925-36