Pasar al contenido principal

Breaking the symmetry in a redox flow battery

IJC2020-044087-I

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de Promoción del Talento y su Empleabilidad en I+D+i
Subprograma Subprograma Estatal de Incorporación
Convocatoria Juan de la Cierva Incorporación
Año convocatoria 2020
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POLITÈCNICA DE VALÈNCIA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 10
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Statistical analysis of the effect of the electrochemical treatment and the acid concentration on the leaching of NMC cathodes from spent Li-ion batteries

  • Grima-Carmena, Laura
  • Bosch-Mossi, Francisco
  • Oyonarte Andrés, Silvia
  • Giner-Sanz, Juan José|||0000-0003-0441-6102
  • García Gabaldón, Montserrat|||0000-0003-4254-6733
  • Pérez-Herranz, Valentín|||0000-0002-4010-0888
[EN] Electric mobility has sharply increased the demand for batteries, making a European priority their sustainability throughout their entire lifecycle. However, battery recycling still needs to be improved, since it relies on industrial processes that require high amounts of energy, are not selective and require chemical reagents such as concentrated acids and reducing agents. In this work, the effect of acid concentration and electrochemical treatment on the leaching of Li, Ni, Mn, Co from electric vehicle LiNixMnyCozO2 (NMC) batteries was studied at room temperature. A factorial 32 experimental design was considered to deconvolve the effect of the chemical leaching from the electrochemical treatment in NMC batteries. An analysis of the variance was performed to determine which factors have a statistically significant effect on the initial leaching rate and the final leaching yield of each one of the metals. The main advantage and novelty of this work is that the combination of the electrochemical treatment and the acid concentration is proved to be significant for the extraction yield of the metals contained in Li-ion batteries of NMC cathode. All the experiments have been done in galvanostatic mode, with no pre-treatments and at room temperature with no addition of reducing agents. Based on the results obtained from the statistical analysis, a novel environmentally friendly process for recycling the cathodes of spent batteries was proposed. This 3-sequencial-step process would allow to recover separately the Al foil, the Li fraction, the Mn fraction, and the Ni-Co fractions., This work was supported by the Instituto Valenciano de Competitividad Empresarial (IVACE) (project reference: IMDEEA/2021/10) and European Union (FEDER funds). GAIKER technology centre assistance is gratefully acknowledged for their work in battery acquisition and dismantling. J.J.G.S. is very grateful to to the Ministerio de Ciencia e Innovacion, to the Next Generation EU, and to the Agencia Estatal de Investigacion, for their support by a Juan de la Cierva-Incorporacion fellowship (IJC2020-044087-I) funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. Authors are thankful to Nuria Fernandez Ortiz for the assistance in ICP- OES characterization of the samples.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Effect of the anode material, applied current and reactor configuration on the atenolol toxicity during an electrooxidation process

  • Montañés, Maria-Teresa|||0000-0002-2620-6926
  • García Gabaldón, Montserrat|||0000-0003-4254-6733
  • Giner-Sanz, Juan José|||0000-0003-0441-6102
  • Mora-Gómez, Julia|||0000-0002-1781-8664
  • Pérez-Herranz, Valentín|||0000-0002-4010-0888
[EN] Atenolol (ATL) is a beta-blocker pharmaceutical product which is excreted mainly unchanged and may represent a long-term risk for organisms present in the sea and in fresh water. Due to its low biodegradation rate, electrochemical advanced oxidation processes (EAOPs) can be used to remove this compound. In this work, ATL ecotoxicity was analyzed in the presence of sodium sulfate (Na2SO4), which is widely used as supporting electrolyte in EAOPs. Ecotoxicity values were expressed as the pollutant concentration that leads to a 50% inhibition of the root elongation of Lactuca sativa seeds in relation to the control (EC50(5 days)). The obtained values for ATL showed an EC50(5 days) of 1377 mg L-1 towards Lactuca sativa. When Na2SO4 was added, the toxicity of the sample increased but no synergy was detected between both compounds. With 2 g L-1 Na2SO4, ATL showed an EC50(5 days) of 972 mg L-1; and with 4 g L-1 Na2SO4 and higher concentrations, EC50 value for ATL was 0 mg L-1. Statistical tools were used to obtain the zones of the [ATL]-[Na2SO4] plane which are toxic towards Lactuca sativa. Solutions containing ATL and Na2SO4 were treated by electrooxidation. Two anode materials (a boron-doped diamond electrode and a microporous Sb-doped SnO2 ceramic one); three operation currents (0.4, 0.6 and 1 A); and two reactor configurations (one-compartment reactor and two-compartment reactor separated by a cation exchange membrane) were used. Lactuca sativa seeds and Vibrio fischeri bacterium tests were employed to evaluate the toxicity of the solutions before and after applying the electrooxidation process. In all the tests, the ecotoxicity of the treated sample increased. This fact is owing to the persulfate presence in the solution due to the sulfate electrochemical oxidation. Nevertheless, none of the final samples were toxic towards Vibrio fischeri because ecotoxicity values were lower than 10 TU; and, in the case of the onecompartment reactor, practically all of them were also non-toxic towards Lactuca sativa. The toxicity of the treated samples increased when using the two-compartment reactor in the presence of the BDD anode, and when the operation current was increased. This is attributed to the highest formation of persulfates. Amongst all the tests performed in this work, the lowest toxicity value (i. e., 3 TU) together with the complete mineralization and degradation degrees was achieved with the two-compartment reactor using the BDD anode and operating at 0.6 A., The authors express their gratitude to the Generalitat Valenciana (GVA-THINKINAZUL/2021/013 Project) and to the Next Generation EU (European Union NextGenerationEU/PRTR) for the financial aid received. J.J.G.S. thanks the Ministerio de Ciencia e Innovacion, the Next Generation EU, and the Agencia Estatal de Investigacion, for their support through a Juan de la CiervaIncor- poracion fellowship (IJC2020-044087-I) funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Effect of the injection and filtration flow rates on the electrofiltration performance of three-dimensional porous Sb-SnO2 anodes

  • Carvalho-Rosa, Germano|||0009-0008-7942-4354
  • Giner-Sanz, Juan José|||0000-0003-0441-6102
  • Pérez-Herranz, Valentín|||0000-0002-4010-0888
  • Mestre, Sergio
[EN] Electrofiltration is an emerging electrochemical technique that enables simultaneous filtration and advanced oxidation for water treatment, offering improved control over mass transport and surface activation compared with conventional electrochemical process. Ceramic 3D Sb-doped SnO2 porous electrodes have been proven to be suitable electrofiltration anodes. In this work, the impact of hydrodynamics on the electrochemical behavior and oxidizing capability of such electrofiltration anodes was investigated. Cyclic voltammetry revealed that, without filtration, the system behaves as a quasi-2D flow-by electrode with low internal resistance, limited electroactive area and low oxygen evolution potential at high injection flow. Under filtration, the system behaves as a 3D flow-through electrode exhibits a higher ohmic resistance for all the tested flow rates, an anodic shift of the oxygen evolution potential and a higher activated area. The ability to generate hydroxyl radicals was assessed by salicylic acid electrofiltration and quantification of 2,5- dihydroxybenzoic acid and 2,3-dihydroxybenzoic acid, confirming that hydroxyl radical production at 20 L/h is effective both, without filtration and with filtration; whereas at 60 L/h it is strongly hindered in flow-by mode and only partially recovered under flow-through operation. These findings highlight the critical role of hydrodynamics in electrofiltration with porous ceramic Sb¿SnO2 anodes and provide a mechanistic link between flow regime, electrochemical response and radical-mediated oxidation performance., The authors are greatly thankful to the Generalitat Valenciana (Project GVA-THINKINAZUL/2021/013), the European Union NextGeneration EU/PRTR. G. Carvalho-Rosa thanks the Generalitat Valenciana (Santiago Grisolía scholarship CIGRIS/2021/037) for funding this research. J.J. Giner-Sanz is very grateful to the Ministerio de Ciencia e Innovacion, to the Next Generation EU, and to the Agencia Estatal de Investigacion, for their support by a Juan de la CiervaIncorporacion ¿ fellowship (IJC2020-044087-I) funded by MCIN/AEI/ 10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. The Article Processing Charges (APCs) were covered by the transformative agreements between CRUE-Universitat Politecnica de Valencia and Elsevier.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Deep Learning algorithm, based on convolutional neural networks, for equivalent electrical circuit recommendation for Electrochemical Impedance Spectroscopy

  • Sáez-Pardo, Fermín|||0009-0000-0759-6781
  • Giner-Sanz, Juan José|||0000-0003-0441-6102
  • Pérez-Herranz, Valentín|||0000-0002-4010-0888
[EN] Electrochemical Impedance Spectroscopy (EIS) is a technique widely used in the field of electrochemistry, due to its ability to probe the dynamics of electrochemical systems. Commonly, EIS spectra are analysed using Equivalent Electrical Circuits (EECs). The EEC selection is not trivial. For this reason, Digby D. Macdonald proposed the Electrochemical Genome Project, which would consist of a large database of EIS spectra and an Artificial Intelligence able to recommend EECs given an experimental EIS spectrum. In this work, we developed a Deep Learning algorithm, based on Convolutional Neural Networks (CNNs), for EEC recommendation for EIS spectra. To achieve this, we optimized the CNN in 3 sequential stages: first, the convolutional architecture was optimized; then, the Initial Learn Rate was selected; and finally, the dense network architecture was optimized. At the end of this process, we obtained a CNN model with a maximum test accuracy of 61.11 %. The obtained results show that CNNs are good candidates for EEC recommendation tools for EIS spectra., This work was funded by an "Ayuda a Primeros Proyectos de Investigacion" project (PAID-06-23) of the research vice-rectorate of Universitat Politecnica de Valencia and the Grant PID2023-147511OB-C21 funded by MICIU/AEI/10.13039/501100011033 and by ERDF/EU. F.S.P. acknowledges the support of Universitat Politecnica de Valencia through a predoctoral fellowship (PAID-01-22) . J.J.G.S. is very grateful to the Ministerio de Ciencia e Innovacion, to the Next Generation EU, and to the Agencia Estatal de Investigacion, for their support by a Juan de la Cierva-Incorporacion fellowship (IJC2020-044087-I) funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. Authors thank Universitat Politecnica de Valencia for funding for open access charge (PAID-12-25) .
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Statistical benchmarking of chloride and sulphate posolytes for all-iron redox flow batteries

  • Cifre-Herrando, Mireia|||0000-0002-8800-3585
  • C. Escoms-Espert
  • Sáez-Pardo, Fermín|||0009-0000-0759-6781
  • Pérez-Herranz, Valentín|||0000-0002-4010-0888
  • Giner-Sanz, Juan José|||0000-0003-0441-6102
[EN] All-iron Redox Flow Batteries (AIRFBs) are promising large-scale energy storage systems due to their low cost and chemical symmetry. However, their performance is largely limited by the slow kinetics of the Fe3+|Fe2+ redox couple in the posolyte, and the posolyte's stability window. A good electrolyte for AIRFBs must combine fast and reversible charge-transfer kinetics, high selectivity toward the Fe3+|Fe2+ reaction, high ionic conductivity, and efficient mass transport. This study systematically investigates how the electrolyte type (chloride and sulphate) and acidity affect the key electrochemical parameters of the Fe3+|Fe2+ redox couple, namely the standard heterogenous rate constant (¿0), the charge transfer coefficient (¿), the diffusion coefficient (¿), and the conductivity (¿). A multifactorial experimental design was employed, and an ANalysis Of VAriance (ANOVA) was used to identify the statistically significant effects of both factors and their interaction on the aforementioned electrochemical parameters. The ANOVA results show a statistically significant difference in the charge-transfer process between chloride- and sulphate-based electrolytes: the electrolyte type primarily controls electron-transfer kinetics, with chloride medium markedly enhancing ¿0; while acid concentration mainly affects ¿ and ¿. Among the tested formulations, 1 M HCl exhibited the best compromise between Fe3+|Fe2+ kinetics, mass transport, and an adequate stability window. These insights provide a quantitative basis for electrolyte optimization and support the rational design of improved AIRFB posolytes., This work was funded by an Ayuda a Primeros Proyectos de Investigacion project (PAID-06 23) of the research vice-rectorate of Universitat Politecnica de Valencia and the Grant PID2023-147511OBC21 funded by MICIU/AEI/10.13039/ 501100011033 997 and by ERDF/EU. M.C.H is grateful to UPV for the post-doctoral grant (PAID10 24). F.S.P. acknowledges the support of Universitat Politecnica de Valencia through a predoctoral fellowship (PAID-01 22). J.J.G.S. is very grateful to the Ministerio de Ciencia e Innovacion, to the Next Generation EU, and to the Agencia Estatal de Investigacion, for their support by a Juan de la Cierva-Incorporacion fellowship (IJC2020-044087-I) funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. The Article Processing Charges (APCs) were covered by the transformative agreements between CRUE Universitat Politecnica de Valencia and Elsevier.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Degradation of multicomponent pharmaceutical mixtures by electrochemical oxidation: Insights about the process evolution at varying applied currents and concentrations of organics and supporting electrolyte

  • Balseviciute, Adele|||0000-0003-0853-5790
  • Giner-Sanz, Juan José|||0000-0003-0441-6102
  • García Gabaldón, Montserrat|||0000-0003-4254-6733
  • Pérez-Herranz, Valentín|||0000-0002-4010-0888
  • Martí Calatayud, Manuel César|||0000-0002-0745-1918
  • Patiño-Cantero, I.
  • Carrillo-Abad, J.
[EN] Electrochemical oxidation is a treatment process that can achieve high degrees of removal of organic emerging pollutants from wastewater. The present study evaluates the removal and mineralization degrees achieved when treating by electrochemical oxidation multicomponent mixtures containing three model pharmaceuticals: atenolol, ibuprofen and norfloxacin. The progress of the removal process at different applied currents was followed through the evolution of the degree of mineralization and the concentration of ionic by-products. The three compounds can be effectively removed, although atenolol was found to be the most recalcitrant one. An increase in the degree of mineralization with current was noted for currents above 200 mA, reaching a maximum value of 98.8 % for an operation current of 800 mA. In general, total organic carbon declines linearly during the first stages of the process, while an exponential trend appears at the latter stages when complete mineralization is approached. Peak concentrations of some by-products, such as formate, ammonia and nitrite ions, were detected at intermediate operating times. Such compounds are further oxidized, thus being depleted from the solution towards the end of the process. On the contrary, a final plateau concentration is reached for fluoride and nitrate ions. Such events occur faster at higher operation currents, thus denoting an accelerated process of mineralization. However, higher currents also induce a decrease in the mineralization current efficiency and an increase in the specific energy consumption. Notably, for a given current, the process evolves at a similar pace regardless of the concentration of organics and supporting electrolyte. In terms of net mineralization, this implies a more efficient utilization of the generated hydroxyl radicals in the removal of organics for higher solution concentrations. Experiments conducted at constant current and organics concentration, but varying the concentration of sodium sulfate, confirmed that hydroxyl radicals are the main species responsible for the removal of pharmaceuticals, whereas the oxidation of the supporting electrolyte only plays a secondary role., The authors thank Conselleria d'Innovacio, Universitats, Ciencia i Societat Digital (project reference: CIGE/2021/173) and the Agencia Estatal de Investigacion for the financial support through the grant (PID2023-147511OB-C21) funded by MICIU/AEI/10.13039/501100011033 and by ERDF/EU. J.J.G.S. is very grateful to the Ministerio de Ciencia e Innovacion, to the Next Generation EU, and the Agencia Estatal de Investigacion for their support by a Juan de la Cierva- Incorporacion fellowship (IJC2020-044087-I) funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. Funding for open access charge: Universitat Politecnica de Valencia.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Study of the effect of the pretreatment and the concentrations of iron (II) and chloride, on the electrochemical leaching of waste printed circuit boards using ANOVA analysis.

  • Grima-Carmena, Laura
  • Pocoví-Martínez, Salvador
  • Bosch-Mossi, Francisco
  • Giner-Sanz, Juan José|||0000-0003-0441-6102
  • Pérez-Herranz, Valentín|||0000-0002-4010-0888
[EN] The current rates of production and consumption of electronic devices have made the generation of electronic waste a global concern. Waste printed circuit boards (WPCBs) are the core component of electrical waste. A proper management of WPCBs is crucial from both, an environmental point of view (they contain toxic materials) and an economic point of view since they contain valuable materials such as precious metals (PMs). The heterogeneity in WPCBs composition hampers the effective recovery of the valuable metals contained in WPCBs, making it essential to adopt a combined approach such as the inclusion of a chemical pretreatment to recover the materials that they contain. This work studies the effect of the chemical pretreatment, Fe2+ and Cl- concentrations on the recovery yield and kinetics of the leaching of Cu, Au and Ag, during the electrochemical treatment from real WPCBs. This study aims to assess the viability of using waste spent pickling acids (SPAs) from surface treatment industries to recover valuable materials from WPCBs by electrochemical leaching. Two full 22 factorial designs were considered: the first design evaluated the effect of Cl¿ and Fe2+ concentrations, while the second design studied the effect of the chemical pretreatment and Cl¿ concentration. A novel in-house leaching system was used. Operating at room temperature with a mildly acidic media (0.05 M H2SO4), the system achieved leaching yields up to 99%, 75% and 99% for Cu, Au, and Ag, respectively. These recoveries were reached in less than 150¿min for Cu and Ag and less than 60 min for Au. Based on the results, a sequential and environmentally friendly recovery process was proposed to recover separately base and PMs from WPCBs., This work was supported by the IVACE (Institut Valencia de Competitivitat Empresarial. Project LIXMIX: IMDEEA/2022/54), and FEDER funds i.e. European Union. Authors are grateful to GAIKER for the WPCBs pretreatment and to Recuintec for the WPCBs. J.J.G.S. is very grateful to the Ministerio de Ciencia e Innovacion, to the Next Generation EU, and to the Agencia Estatal de Investigacion, for their support by a Juan de la Cierva-Incorporacion fellowship (IJC2020 044087-I) funded by MCIN/AEI/10.13039/ 501100011033 and by the European Union NextGenerationEU/PRTR. The authors express their gratitude to Nuria Fernández Ortiz for the support provided in characterizing the samples. The Article Processing Charges (APCs) were covered by the transformative agreements between CRUE-Universitat Politecnica de Valencia and Elsevier.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Statistical comparison of the photoelectrochemical degradation of an antibiotic pollutant using two Sb-doped SnO2 ceramic anodes coated with photoactive CdFe2O4 and ZnFe2O4 layers

  • Medina-Casas, Martha Patricia|||0000-0003-4503-432X
  • Pérez-Herranz, Valentín|||0000-0002-4010-0888
  • Giner-Sanz, Juan José|||0000-0003-0441-6102
  • García Gabaldón, Montserrat|||0000-0003-4254-6733
  • Mestre, Sergio
[EN] In the present study, the electrochemical and photoelectrochemical responses of two Sb-doped SnO2 ceramic anodes were compared in the degradation of an emerging pharmaceutical pollutant, the antibiotic Norfloxacin. One anode was coated with a photocatalytic layer of cadmium ferrite (CdFe2O4), while the other one was coated with zinc ferrite (ZnFe2O4). A multifactorial analysis of variance (ANOVA) was used to assess the effect of the anode material, the applied current intensity and the presence or absence of light, on the degradation and mineralization kinetics. An initial electrochemical characterization of the electrodes was conducted using linear sweep voltammetry and light-pulsed chronoamperometry, confirming the photoactivity of these electrodes. The results revealed that at intermediate potential values (around 1.8 V vs. Ag/AgCl 3 M for cadmium ferrite anode and 2.4 V vs. Ag/AgCl 3 M for zinc ferrite anode), the largest photoactivity were observed. The degradation rate of Norfloxacin was positively influenced by the presence of light and by the increase in current intensity. Additionally, the combined effect of the electrode material and light presence resulted in higher final degradation values of the compound. The degradation kinetics of Norfloxacin were fitted using a mixed-order model (pseudo-zero order and pseudo-first order). The pseudo-zero order kinetic constant was positively influenced by all the considered factors and by the interaction between electrode material and light, and between current intensity and light, which resulted in higher constant values. All the studied factors had a positive effect on Norfloxacin mineralization. Additionally, the combined effect of the electrode material and current intensity showed a significant improvement in the final mineralization of the compound. Furthermore, it was confirmed that the mineralization current efficiency (MCE) decreases with increasing applied current intensity, but this decrease is lower when the electrode is exposed to light., The authors are greatly thankful to the Generalitat Valenciana (Project GVA-THINKINAZUL/2021/013), the European Union NextGeneration EU/PRTR. M. P. Medina-Casas thanks the Generalitat Valenciana (Santiago Grisolía scholarship CIGRIS/2021/035) for funding this research. J.J. Giner-Sanz is very grateful to the Ministerio de Ciencia e Innovacion, ¿ to the Next Generation EU, and to the Agencia Estatal de Investigacion, ¿ for their support by a Juan de la CiervaIncorporacion ¿ fellowship IJC2020-044087-I funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Modeling of a solution polymerization reactor operating in semi-batch mode for polymethyl methacrylate production

  • Sáez-Pardo, Fermín|||0009-0000-0759-6781
  • Giner-Sanz, Juan José|||0000-0003-0441-6102
  • García Gabaldón, Montserrat|||0000-0003-4254-6733
  • Pérez-Herranz, Valentín|||0000-0002-4010-0888
[EN] The severe consequences of thermal runaways, make process optimization of paramount importance for free radical polymerization reactors, in order to maximize in an integrated way their safety and productivity. Generally, this optimization is performed by simulation, for the sake of safety and cost. Today, the literature contains several models of free radical polymerization reactors in general; and methyl methacrylate (MMA) polymerization reactors, in particular. Although, MMA polymerization is commonly performed in semi-batch reactors at industrial level, most of the available models focus on batch reactors; while, the semi-batch reactor models are much scarcer. In this work, a model of a MMA solution polymerization batch reactor was modified in order to generalize it to semi-batch operation. The developed semi-batch model was used here for studying the effect of different operation conditions (feeding flow rate, initiator load, monomer initial load, and reactor temperature) on the quality characteristics of the produced polymer and on the safety and productivity of the polymerization reactor. The developed model can be used to optimize the operation conditions of a semibatch reactor so that the final product quality meets the application requirements, while maintaining the reactor within its safe-operation envelope, and maximizing its productivity., F.S.P. acknowledges the support of Universitat Politecnica de Valencia through a predoctoral fellowship (PAID-01-22) . J.J.G.S. is very grateful to the Ministerio de Ciencia e Innovacion, to the Next Generation EU, and to the Agencia Estatal de Investigacion, for their support by a Juan de la Cierva-Incorporacio n fellowship IJC2020-044087-I funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. Funding for open access was provided by CRUE-Universitat Politecnica de Valencia.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Equivalent Electrical Circuit recommendation for Electrochemical Impedance Spectroscopy: A benchmark of different Machine Learning algorithms

  • Sáez-Pardo, Fermín|||0009-0000-0759-6781
  • Giner-Sanz, Juan José|||0000-0003-0441-6102
  • Pérez-Herranz, Valentín|||0000-0002-4010-0888
[EN] Electrochemical Impedance Spectroscopy (EIS) is a technique widely used in the electrochemistry field that allows assessing the kinetic parameters of electrochemical systems. In general, EIS spectra are analyzed using 2 methodologies: the physical¿mathematical methodology and the Equivalent Electrical Circuit (EEC) methodology. The physical¿mathematical methodology consists in developing and solving the set of differential equations that govern the system. In contrast, the EEC methodology describes EIS data in terms of EEC models, decreasing the math complexity and democratizing the use of the EIS technique. In the EEC methodology, the most critical task is selecting a physically sound EEC model. To help with the EEC model selection, Prof. Digby D. Macdonald proposed creating the Electrochemistry¿s Genome Project, which would consist of a large database and an Artificial Intelligence (AI) able to suggest plausible EEC models given an EIS spectrum. Until now, several works have explored the idea of using Machine Learning (ML) for recommending an EEC for a given EIS spectrum. Indeed, several authors have benchmarked different ML algorithms for this task. However, these benchmarks generally report point estimates of the accuracy achieved by each algorithm, but do not assess its uncertainty. In this work, we benchmarked different ML EEC recommendation algorithms. To achieve this, we refined a preexistent database, obtaining a new database. Then, the new database was used to train different ML algorithms, optimize their hyperparameters, and assess their accuracy. Finally, the performance of the different optimized algorithms were compared. Whereas most of the other benchmarks available in literature work with point estimates of the accuracies, in this work, the accuracy distributions have been estimated. This methodology allows to compare the performance of the different algorithms in a much more reliable and robust way., F.S.P. acknowledges the support of Universitat Politecnica de Valencia through a predoctoral fellowship (PAID-01-22). J.J.G.S. is very grateful to the Ministerio de Ciencia e Innovacion, to the Next Generation EU, and to the Agencia Estatal de Investigacion, for their support by a Juan de la Cierva-Incorporacion fellowship IJC2020-044087-I funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR.