Automatically Identifying Electrode Reaction Mechanisms Using Deep Neural Networks.

At present, electrochemical mechanisms are most commonly identified subjectively based on the experience of the researcher. This subjectivity is reflected in bias to particular mechanisms as well as lack of quantifiable confidence in the chosen mechanism compared to potential alternative mechanisms. In this paper we demonstrate that a deep neural network trained to recognize dc cyclic voltammograms for three commonly encountered mechanisms provides correct classifications within 5 ms without the problem of subjectivity. To mimic experimental data, the impact of noise, uncompensated resistance, and dependence on scan rate, factors that are relevant to practical studies, has also been investigated. Outcomes with two experimental data sets are also presented.

Application of Bayesian Inference in Fourier-Transformed Alternating Current Voltammetry for Electrode Kinetic Mechanism Distinction.

Estimation of parameters of interest in dynamic electrochemical (voltammetric) studies is usually undertaken via heuristic or data optimization comparison of the experimental results with theory based on a model chosen to mimic the experiment. Typically, only single point parameter values are obtained via either of these strategies without error estimates. In this article, Bayesian inference is introduced to Fourier-transformed alternating current voltammetry (FTACV) data analysis to distinguish electrode kinetic mechanisms (reversible or quasi-reversible, Butler-Volmer or Marcus-Hush models) and quantify the errors. Comparisons between experimental and simulated data were conducted across all harmonics using public domain freeware (MECSim).

Access to enhanced differences in Marcus-Hush and Butler-Volmer electron transfer theories by systematic analysis of higher order AC harmonics.

The potential-dependences of the rate constants associated with heterogeneous electron transfer predicted by the empirically based Butler-Volmer and fundamentally based Marcus-Hush formalisms are well documented for dc cyclic voltammetry. However, differences are often subtle, so, presumably on the basis of simplicity, the Butler-Volmer method is generally employed in theoretical-experimental comparisons. In this study, the ability of Large Amplitude Fourier Transform AC Cyclic Voltammetry to distinguish the difference in behaviour predicted by the two formalisms has been investigated. The focus of this investigation is on the difference in the profiles of the first to sixth harmonics, which are readily accessible when a large amplitude of the applied ac potential is employed. In particular, it is demonstrated that systematic analysis of the higher order harmonic responses in suitable kinetic regimes provides predicted deviations of Marcus-Hush from Butler-Volmer behaviour to be established from a single experiment under conditions where the background charging current is minimal.

Theoretical analysis of the two-electron transfer reaction and experimental studies with surface-confined cytochrome c peroxidase using large-amplitude Fourier transformed AC voltammetry.

A detailed analysis of the cooperative two-electron transfer of surface-confined cytochrome c peroxidase (CcP) in contact with pH 6.0 phosphate buffer solution has been undertaken. This investigation is prompted by the prospect of achieving a richer understanding of this biologically important system via the employment of kinetically sensitive, but background devoid, higher harmonic components available in the large-amplitude Fourier transform ac voltammetric method. Data obtained from the conventional dc cyclic voltammetric method are also provided for comparison. Theoretical considerations based on both ac and dc approaches are presented for cases where reversible or quasi-reversible cooperative two-electron transfer involves variation in the separation of their reversible potentials, including potential inversion (as described previously for solution phase studies), and reversibility of the electrode processes. Comparison is also made with respect to the case of a simultaneous two-electron transfer process that is unlikely to occur in the physiological situation. Theoretical analysis confirms that the ac higher harmonic components provide greater sensitivity to the various mechanistic nuances that can arise in two-electron surface-confined processes. Experimentally, the ac perturbation with amplitude and frequency of 200 mV and 3.88 Hz, respectively, was employed to detect the electron transfer when CcP is confined to the surface of a graphite electrode. Simulations based on cooperative two-electron transfer with the employment of reversible potentials of 0.745 ± 0.010 V, heterogeneous electron transfer rate constants of between 3 and 10 s(-1) and charge transfer coefficients of 0.5 for both processes fitted experimental data for the fifth to eighth ac harmonics. Imperfections in theory-experiment comparison are consistent with kinetic and thermodynamic dispersion and other nonidealities not included in the theory used to model the voltammetry of surface-confined CcP.

Large amplitude Fourier transformed ac voltammetry at a rotating disc electrode: a versatile technique for covering Levich and flow rate insensitive regimes in a single experiment.

The theory for large amplitude Fourier transformed ac voltammetry at a rotating disc electrode is described. Resolution of time domain data into dc and ac harmonic components reveals that the mass transport for the dc component is controlled by convective-diffusion, while the background free higher order harmonic components are flow rate insensitive and mainly governed by linear diffusion. Thus, remarkable versatility is available; Levich behaviour of the dc component limiting current provides diffusion coefficient values and access to higher harmonics allows fast electrode kinetics to be probed. Two series of experiments (dc and ac voltammetry) have been required to extract these parameters; here large amplitude ac voltammetry with RDE methodology is used to demonstrate that kinetics and diffusion coefficient information can be extracted from a single experiment. To demonstrate the power of this approach, theoretical and experimental comparisons of data obtained for the reversible [Ru(NH(3))(6)](3+/2+) and quasi-reversible [Fe(CN)(6)](3-/4-) electron transfer processes are presented over a wide range of electrode rotation rates and with different concentrations and electrode materials. Excellent agreement of experimental and simulated data is achieved, which allows parameters such as electron transfer rate, diffusion coefficient, uncompensated resistance and others to be determined using a strategically applied approach that takes into account the different levels of sensitivity of each parameter to the dc or the ac harmonic.

Leveraging e-Science infrastructure for electrochemical research.

As in many scientific disciplines, modern chemistry involves a mix of experimentation and computer-supported theory. Historically, these skills have been provided by different groups, and range from traditional 'wet' laboratory science to advanced numerical simulation. Increasingly, progress is made by global collaborations, in which new theory may be developed in one part of the world and applied and tested in the laboratory elsewhere. e-Science, or cyber-infrastructure, underpins such collaborations by providing a unified platform for accessing scientific instruments, computers and data archives, and collaboration tools. In this paper we discuss the application of advanced e-Science software tools to electrochemistry research performed in three different laboratories--two at Monash University in Australia and one at the University of Oxford in the UK. We show that software tools that were originally developed for a range of application domains can be applied to electrochemical problems, in particular Fourier voltammetry. Moreover, we show that, by replacing ad-hoc manual processes with e-Science tools, we obtain more accurate solutions automatically.

Effects of coupled homogeneous chemical reactions on the response of large-amplitude AC voltammetry: extraction of kinetic and mechanistic information by Fourier transform analysis of higher harmonic data.

Large-amplitude ac voltammograms contain a wealth of kinetic information concerning electrode processes and can provide unique mechanistic insights compared to other techniques. This paper describes the effects homogeneous chemical processes have on ac voltammetry in general and provides experimental examples using two well-known chemical systems: one simple and one complex. Oxidation of [Cp*Fe(CO)(2)](2) (Cp* = η(5)-pentamethylcyclopentadienyl) in noncoordinating media is a reversible one-electron process; in the presence of nucleophiles, however, the resulting ligand-induced disproportionation changes the process to a multiple step regeneration. The chemical kinetic parameters of the regeneration mechanism were discerned via analysis of the third and higher harmonics of Fourier-transformed ac voltammetry data. Comparison of experimental data to digital simulations provides clear evidence that the reaction proceeds via a rapid pre-equilibrium between the electrogenerated monocation and the coordinating ligand; simultaneous fitting of the first nine harmonics indicates that k(f) = 7500 M(-1) s(-1) and k(r) = 100 s(-1), and that the unimolecular decomposition of the corresponding intermediate occurs with a rate constant of 2.2 s(-1). The rapid cis(+) → trans(+) isomerization of the electrogenerated cis-[W(CO)(2)(dpe)(2)](+), where dpe = 1,2-diphenylphosphinoethane, was examined to illustrate the effects of a simpler EC mechanism on the higher harmonics; a rate constant of 280 s(-1) was determined. These results not only shed new light on the chemistry of these systems, but provide a clear demonstration that the higher harmonics of ac voltammetry provide mechanistic insights into coupled homogeneous processes far more detailed than those that are readily accessible with dc techniques.