The deep-DRT: A deep neural network approach to deconvolve the distribution of relaxation times from multidimensional electrochemical impedance spectroscopy data

Quattrocchi, Emanuele; Wan, Ting Hei; Belotti, Alessio; Kim, Dohyung; Pepe, Simona; Kalinin, Sergei V.; Ahmadi, Mahshid; Ciucci, Francesco

doi:10.1016/j.electacta.2021.139010

Cited by 71 publications

(43 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also studied EIS data measured from a perovskite solar cell. The cell was made of fluorinedoped tin oxide, the electron-transport layer of SnO 2 , the transport-hole layer of Spiro-OMeTAD, and the absorber of (FAPbI 3 ) 0.85 (MAPbBr 3 ) 0.15 [23]. The EIS spectrum was measured from 1 Hz to 5 MHz in a 1 atm atmosphere with a 100 mV voltage amplitude.…”

Section: Perovskite Solar Cellmentioning

confidence: 99%

“…EIS is particularly useful because it enables the characterization of such systems to obtain important parameters like conductivities [16,17], resistances [18,19], and capacitances [20,21]. However, issues remain, as the determination of these parameters may depend closely on the equivalent circuit model (ECM) used to deconvolve the EIS spectrum [22][23][24]. In addition, EIS data are typically affected by experimental errors, making interpolation and prediction of the spectra at untested frequencies challenging [4,25].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Gaussian Processes for the Analysis of Electrochemical Impedance Spectroscopy Data: Prediction, Filtering, and Active Learning

Maradesa

Ciucci

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Electrochemical impedance spectroscopy (EIS) is a widespread characterization technique used to study electrochemical systems. However, several shortcomings still limit the application of this technique. First, EIS data is intrinsically noisy, hindering spectra regression and prediction at unknown frequencies. Second, many physicochemical properties, such as the polarization resistance, are determined through non-unique equivalent circuits. Third, probed frequencies are usually log-spaced with a fixed number of points per decade, which is not necessarily optimal. This article illustrates how Gaussian processes (GPs) can overcome these three issues by showing that GPs can successfully filter EIS data and be used to determine the polarization resistance as a stochastic variable. Lastly, a GP-based, active-learning framework is developed to select EIS frequencies optimally for quick and accurate measurements.

show abstract

Section: Perovskite Solar Cellmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Gaussian Processes for the Analysis of Electrochemical Impedance Spectroscopy Data: Prediction, Filtering, and Active Learning

Maradesa

Ciucci

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…It was recently shown that deep neural networks (DNNs) can be utilized to deconvolve the DRT [50,51]. In particular, the first DNN-based method ever developed, called deep-prior DRT, used a DNN with a single random scalar as input [50] and output a vector of DRT values at discrete timescales together with the circuit parameters 𝐿 and 𝑅 .…”

Section: Introductionmentioning

confidence: 99%

“…Quattrocchi et al further extended the deep-prior DRT approach with the deep-DRT model by considering a DNN that takes as inputs the scalar log-timescale log 𝜏 and a state vector 𝝍 = (𝜓 , … , 𝜓 ) describing experimental conditions (e.g., temperature, pressure, etc.) [51]. In addition to inverting the DRT and regressing the experimental impedance, the trained deep-DRT model was used to predict the DRT (and corresponding impedance) at experimental states not tested experimentally.…”

Section: Introductionmentioning

confidence: 99%

“…Even though the DNN is a universal approximator that closely matches any function (see Section S1 of the supplementary information (SI)), DRT deconvolution has been performed with a limited accuracy because the integral in (1) needs to be approximated at a fixed number of collocation points [47,50]. iii) The recovery of DRTs with negative peaks has yet to be studied as the output of DNN methods that have been constrained to be non-negative by means of a dedicated activation function at the DNN's last layer [50,51]. This article aims to address thesechallenges.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deconvolution of Electrochemical Impedance Spectroscopy Data Using the Deep-Neural-Network-Enhanced Distribution of Relaxation Times

Quattrocchi

Maradesa

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Electrochemical impedance spectroscopy (EIS) is a characterization technique widely used to evaluate the properties of electrochemical systems. The distribution of relaxation times (DRT) has emerged as a model-free alternative to equivalent circuits and physical models to circumvent the inherent challenges of EIS analysis. Deep neural networks (DNNs) can be used to deconvolve DRTs, but several issues remain, e.g., the long training time, the DNN accuracy, and the deconvolution of DRTs with negative peaks. The DNN-DRT model was developed here to address these fundamental limitations. Specifically, a pretraining step was included to decrease the computation time. A thorough error analysis was also conducted to evaluate the different components of the DRT and impedance errors to ultimately decrease them. Lastly, the training loss function was modified to handle DRTs with negative peaks. These different advances were validated with an array of synthetic EIS spectra and real EIS spectra from a lithium-ion battery, a solid oxide fuel cell, and a proton exchange membrane fuel cell. Moreover, this new model outperformed in most cases the previously developed DRTtools and deep-DRT model. Overall, we envision that this research will open the venue for more DNN-based analyses of EIS data for electrochemical systems.

show abstract

A new Co‐based cathode with high performance for intermediate‐temperature solid oxide fuel cells

Zhou,

Liang,

Qiu

et al. 2024

Asia-Pacific J Chem Eng

View full text Add to dashboard Cite

Solid oxide fuel cells (SOFCs) as highly effective energy conversation devices have gained substantial recognition and research interest. The electrochemical properties of the traditional SOFCs are restricted by the sluggish reaction kinetics for the cathode material when lowering the operation temperature to below 600°C. In addition, the stability of the cathode at reduced temperatures is also a big challenge for the widely popularization of SOFC technology. Achieving high activities and stable ORR in the cathode is crucial for the development of SOFCs. The doping active metal method has been demonstrated as an effective approach to optimize the phase structure and improve the ORR activity of the cathode. Herein, we successfully develop an Ir‐doped SrCoO3 − δ (SrCo0.98Ir0.02O3 − δ, SCI) cathode for SOFCs. SCI exhibits a low area‐specific resistance (ASR) of 0.057 Ω cm2 at 650°C, ~ 44% lower than 0.102 Ω cm2 of Ir‐free SrCoO3 − δ. The Ni–Sm0.2Ce0.8O1.90 (SDC) anode‐supported fuel cell with SDC electrolyte and SCI cathode obtains an excellent output performance (e.g., 1,128 mW cm−2 at 650°C). The desired results underscore the feasibility of the Ir‐doping strategy as an optimized method for the exploitation of advancing cathode in SOFCs.

show abstract

The deep-DRT: A deep neural network approach to deconvolve the distribution of relaxation times from multidimensional electrochemical impedance spectroscopy data

Cited by 71 publications

References 59 publications

Gaussian Processes for the Analysis of Electrochemical Impedance Spectroscopy Data: Prediction, Filtering, and Active Learning

Gaussian Processes for the Analysis of Electrochemical Impedance Spectroscopy Data: Prediction, Filtering, and Active Learning

Deconvolution of Electrochemical Impedance Spectroscopy Data Using the Deep-Neural-Network-Enhanced Distribution of Relaxation Times

A new Co‐based cathode with high performance for intermediate‐temperature solid oxide fuel cells

Contact Info

Product

Resources

About