LAUSR

When Fritz Scholz asked me to write an opinion article on ‘Future tasks of electrochemical research’, on a topic of my choice, he also asked me to address the questions: ‘What needs to be done and why?’ and ‘What is missing in present research?’ In response to this daunting challenge, I decided initially to raise for debate a perception I have had for several years:Why is there a significant decline rather than increase of quantitative reporting of voltammetric data, in an era of ready access to supercomputing power, vast arrays of data analysis software and abundant readily accessible theory that can address very complex problems? I needed also to query how my perception could possibly be correct, when machine learning and artificial intelligence have made amazing advances in many branches of science. The impact of advanced data analysis is so high that the Bayesian Statistics method has been the subject of feature articles in major newspapers. Quotes from F.D. Flam in the New York Times published on September 30, 2014, and entitled ‘The Odds, Continually Updated’ provides context: ‘...researchers are using Bayesian statistics to tackle problems of formidable complexity. The New York University astrophysicist David Hogg credits Bayesian statistics with narrowing down the age of the universe...from eight billion to 15 billion years...to 13.8 billion years and when combinedwith advanced computing power has revolutionized the search for planets orbiting distant stars....Bayesian statistics are rippling through everything from physics to cancer research, ecology to psychology ....allowing scientists to solve problems that would have been considered impossible just 20 years ago’. Dynamic voltammetry is a natural technique for implementing data analysis. Why has not the impact of Bayesian Statistics in voltammetry, also not produced a series of groundbreaking achievements in parameterisation of the complex reaction schemes frequently proposed in the modern era of the discipline? To understandwhat needs to be done and why, and what is missing in quantitative voltammetric research in data analysis as requested by Fritz Scholz, I conclude my opinion article by proposing a strategy to facilitate the return of parameterisation of postulated electrochemical mechanisms with error analysis, as the rule, rather than the exception, in the next generation of voltammetric research. My perception that there has been a transition from an expectation of routine quantitative parameterisation of electrode processes in voltammetry, to the present paucity of quantitative data comparisons of experimental data with theoretical predictions, arose from reflecting on the almost universally qualitative content of lectures and posters presented at conferences I have recently attended. I also noted that these conferences were almost completely devoid of studies of any kind, at the highly homogenous and readily modelled mercury electrodes that dominated quantitative voltammetry (polarography) for a significant part of the twentieth century. It was also apparent that conference proceedings were starting to emphasise empirical cyclic voltammetric electrocatalytic studies on carbon dioxide reduction and oxygen or hydrogen evolution at an extraordinary variety of highly heterogeneous electrode-solution interfaces emerging from advances in materials science. In the contributions related to carbon dioxide reduction at conferences, there are always a few examples of use of Tafel Analysis [1]. This form of data analysis is used to provide mechanistic conclusions for multi-step electrocatalytic reactions, as apply in carbon dioxide reduction [2] and is based on slopes of selected sections of ‘log plots’ obtained from a voltammogram. Overpotentials are also often reported in these analyses from use of the onset potential where faradaic current * Alan M. Bond [email protected]