MSE3: Computational electrochemistry
| Researcher: | Graham Morris |
| Team Leader(s): | Dr Kathryn Gillow, Prof. David Gavaghan & Dr Ruth Baker |
| Collaborators: | Prof. Fraser Armstrong |
| Prof. Alan Bond, Monash University |
Background
The computational electrochemistry project aims to extend
the theory for the electroanalytical technique of voltammetry to allow for more
complicated applied potential waveforms. The addition of a sum of sine waves to
the standard cyclic linear ramp allows for investigation of chemical reaction
systems on multiple time scales within a single experiment – a significant
advantage over the commonly used DC method.
Techniques and Challenges
The project can be broken down into two main sections: the forward and the inverse problems. For the forward problem, we make use of techniques in mathematical modelling and numerical analysis to produce and solve models, typically systems of partial or ordinary differential equations (PDEs or ODEs), representing various electron transfer reactions. We then use these models as the framework for an inverse problem, namely the recovery of reaction parameters from a data set representing current response measured in a voltammetric experiment. This involves optimisation methods and the use of Fourier transforms, as we look to deal with the presence of both measurement error and a sometimes significant Background current.
Results
We have produced models for multiple reaction systems, including both those in a solution phase and those confined to a surface, all of which show good correspondence with experimental results. Our inverse problem algorithm in simulation recovers the system parameters to a reasonable degree of accuracy and is often able to determine the reaction mechanism without prior knowledge. When run on actual experimental data, it has so far performed successfully.
The Future
We are extending our models to include experiments run with a rotating disk electrode. The inverse problem algorithm still requires some refinement and further testing on experimental data. We also hope to model kinetic dispersion – an effect that has been noticed by our experimental collaborators.
