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

Electrochemical impedance spectroscopy (EIS) is an experimental technique ubiquitously used to study electrochemical systems. However, conventional EIS data interpretation through physical and equivalent circuit models challenging because are problem-specific, circuits often just lumped-element analogs lacking meaning. The distribution of relaxation times (DRT) has emerged as a complementary approach resolve these issues. One drawback DRT deconvolution that the understood be (only) function frequency (i.e. 1D data) deconvolved accordingly. This work proposes novel method based on deep neural networks (DNNs), allowing analysis multidimensional spectra bridge dependency both frequencies conditions. Two particularly appealing traits deep-DRT developed in this article neither regularization nor specific spacing state variables defining experiment required. Leveraging DNN examine complex their dependence conditions, opens new research direction area deconvolution.