Abstract
The Hopfield Neural Network has been successfully applied to solve ill-posed inverse problems in different fields of chemistry and physics. In this work, the non-linear approach for this method will be applied to retrieve the empirical parameters of potential energy function, \(E_{p}(r)\), between adsorbate and adsorbent from experimental data. Since the adsorption data is related to the second virial coefficient and therefore to \(E_{p}(r)\) through an integral equation, the Hopfield Neural Network will be used to find the best parameters which fits the experimental data. Initially simulated results will be analyzed to verify the method performance for data sets with and without noise addition. Then, experimental data for adsorption of propionitrile on activated carbon will be treated. Results presented here corroborate to the robustness of this method.
Similar content being viewed by others
Data availability
The data sets and codes generated during this work are available from the corresponding author on reasonable request.
References
Moon CJ, Lee JH (2005) Use of curdlan and activated carbon composed adsorbents for heavy metal removal. Process Biochem 40:1279–1283
De Gisi S, Lofrano G, Grassi M, Notarnicola M (2016) Characteristics and adsorption capacities of low-cost sorbents for wastewater treatment: a review. Sustainable Mater 9:10–40
Jeirani Z, Niu CH, Soltan J (2017) Adsorption of emerging pollutants on activated carbon. Rev Chem Eng 33:491–522
Rajahmundry GK, Garlapati C, Kumar PS, Alwi RS, Vo DVN (2021) Statistical analysis of adsorption isotherm models and its appropriate selection. Chemosphere 276:130176
Ghosal PS, Gupta AK (2017) Development of a generalized adsorption isotherm model at solid-liquid interface: A novel approach. J Mol Liq 240:21–24
Ranjan P, Verma P, Agrawal S, Rao TR, Samanta SK, Thakur AD (2019) Inducing dye-selectivity in graphene oxide for cationic dye separation applications. Mater Chem Phys 226:350–355
Carr R, Comer J, Ginsberg MD, Aksimentiev A (2011) Microscopic perspective on the adsorption isotherm of a heterogeneous surface. J Phys Chem Lett 2:1804–1807
Khalfaoui M, Knani S, Hachicha MA, Lamine AB (2003) New theoretical expressions for the five adsorption type isotherms classified by BET based on statistical physics treatment. J Colloid Interface Sci 263:350–356
Nakhli A, Bergaoui M, Aguir C, Khalfaoui M, M’henni MF, Lamine AB, (2016) Adsorption thermodynamics in the framework of the statistical physics formalism: basic blue 41 adsorption onto Posidonia biomass. Desalin Water Treat 57:12730–12742
Torkia YB, Atrous M, Bouzid M, Dotto GL, Lamine AB (2020) Stereographic and energetic studies of acid blue 9 adsorption onto Spirulina platensis (strain LEB-52) based on statistical physics approach. Chem Eng Commun 207:445–457
Oueslati K, Lima EC, Ayachi F, Cunha MR, Lamine AB (2020) Modeling the removal of Reactive Red 120 dye from aqueous effluents by activated carbon. Water Sci Technol 82:651–662
Radke CJ, Prausnitz JM (1972) Statistical mechanics of adsorption from dilute liquid solution. J Chem Phys 57:714–722
Braga JP (2021) Termodinâmica estatística de átomos e moléculas. Livraria da Física, São Paulo
Hadamard J (1923) Lectures on Cauchy’s problem in linear partial differential equations. Yale University Press, New Haven
Leon SJ, Bica I, Hohn T (2006) Linear algebra with applications. Pearson Prentice Hall, Upper Saddle River, NJ
Tikhonov AN, Arsenin VY (1977) Solutions of Ill-Posed Problems. Wiley, New York
Hopfield JJ (1982) Neural networks and physical systems with emergent collective computational abilities. Proc Natl Acad Sci 79:2554–2558
Hopfield JJ, Tank DW (1985) Neuralcomputation of decisions in optimization problems. Biol Cybern 52:141–152
Carvalho FS, Braga JP (2020) Indirect Solution of Ornstein-Zernike Equation Using the Hopfield Neural Network Method. Braz J Phys 50:489–494
Carvalho FS, Braga JP (2020) Radial distribution function for liquid gallium from experimental structure factor: a Hopfield neural network approach. J Mol Model 26:1–5
Carvalho FS, Braga JP, Alves MO, Gonçalves CEM (2020) Neural network in the inverse problem of liquid argon structure factor: from gas-to-liquid radial distribution function. Theor Chem Acc 139:1–6
Sebastião RCO, Braga JP (2005) Retrieval of transverse relaxation time distribution from spin-echo data by recurrent neural network. J Magn Reson 177:146–151
Braga JP, de Almeida MB, Braga AP, Belchior JC (2000) Hopfield neural network model for calculating the potential energy function from second virial data. Chem Phys 260:347–352
Lemes NHT, Borges E, Sousa RV, Braga JP (2008) Potential energy function from differential cross-section data: An inverse quantum scattering theory approach. Int J Quantum Chem 108:2623–2627
Araujo BC, Carvalho FS, Marques MBF, Braga JP, Sebastião RCO (2020) Hopfield Neural Network-Based Algorithm Applied to Differential Scanning Calorimetry Data for Kinetic Studies in Polymorphic Conversion. J Braz Chem Soc 31:1392–1400
Viterbo VC, Braga JP, Braga AP, de Almeida MB (2001) Inversion of simulated positron annihilation lifetime spectrum using a neural network. J Chem Inf Comput Sci 41:309–313
Vemuri V, Jang GS (1992) Inversion of Fredholm integral equations of the first kind with fully connected neural networks. J Franklin Inst 329:241–257
Lemes NHT, Borges E, Braga JP (2007) A general algorithm to solve linear and nonlinear inverse problems. J Braz Chem Soc 18:1342–1347
McLachlan AD (1965) Effect of the medium on dispersion forces in liquids. Discuss. Faraday Soc 40:239–245
Kestner NR, Sinanoǧlu O (1965) Intermolecular forces in dense media. Discuss Faraday Soc 40:266–267
Shampine LF, Reichelt MW (1997) The MATLAB ODE Suite. SIAM J Sci Comput 18:1–22
Shampine LF, Reichelt MW, Kierzenka JA (1999) Solving Index-1 DAEs in MATLAB and Simulink. SIAM Review 41:538–552
Funding
This work received financial support from CNPq.
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Felipe Silva Carvalho and Márcio Oliveira Alves. The code used was developed by João Pedro Braga and adapted to this work by Felipe Silva Carvalho. The first draft of the manuscript was written by Felipe Silva Carvalho and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
This article belongs to the Topical Collection: XXI-Brazilian Symposium of Theoretical Chemistry (SBQT2021)
Rights and permissions
Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Carvalho, F.S., Braga, J.P. & Alves, M.O. Adsorbate-adsorbent potential energy function from second virial coefficient data: a non-linear Hopfield Neural Network approach. J Mol Model 28, 286 (2022). https://doi.org/10.1007/s00894-022-05274-w
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00894-022-05274-w