Multivariate Data Analysis of Multisine Impedimetric Fingerprints in Electroanalysis of Biochemical Compounds

Abstract

While Electrochemical Impedance Spectroscopy (EIS) is found as one of the most versatile tools of electrochemical characterization, it suffers a major drawback when the process in question is non-stationary. On the other hand, Dynamic Electrochemical Impedance Spectroscopy (DEIS) offers an ultra-fast measurement time. DEIS utilizes an approach where all the elementary components are superimposed as one multi-component perturbation signal, offering impedance spectra collection in seconds. This particular advantage allows for the effective use of DEIS in the monitoring of the adsorption processes in multiple fields, such as (bio)sensing, corrosion science, energy storage, or AFM-coupled impedance mapping.This presentation aims at highlighting the DEIS monitoring utility carried out with changes of an independent variable, such as adsorbate concentration or even under linear polarization sweep. Rapid and accurate determination of adsorption isotherms for green organic corrosion inhibitors or Michaelis-Menten constant for enzyme-based glucose (bio)sensors with DEIS operating in galvanostatic mode and under iDC = 0 were already established. When it comes to impedimetric macromolecular electroanalysis, measuring charge transfer resistance upon analyte adsorption is the most common approach. However, the detection may also base on other factors, such as adsorbed layer capacitance or even disturbance in frequency dispersion of capacitance. Most recently, DEIS under polarization sweep mode allowed us to obtain the unique fingerprint of the electric parameter changes, specific to studied macromolecules (such as DNA strings, viral proteins, RNA polymerase, and enzymes). To effectively use our approach DEIS is combined with multivariate data analysis (principal component analysis with partial least square regression). Multivariate data analysis processes large amounts of generated impedimetric data and defines analyte-sensitive conditions (DC polarization, AC amplitude, frequency range, etc.), which can be done just by using the raw data by the singular-value decomposition method to support the sensing mechanisms. The proposed approach neglects some of the reproducibility issues induced by non-specific adsorption and fouling and reveals the measurement conditions that offer the highest variation of studied parameters upon specific analyte adsorption. The authors acknowledge the financial support of these studies by the National Science Centre (Republic of Poland) under project 2020/37/B/ST7/03262.