Introduction

The use of spectroscopic techniques to investigate the structures of and reactions at the solid-liquid interface in electrochemical systems is an area of rapid progress [1], However, because many conventional surface techniques are difficult to apply in these systems directly, there is a growing need to develop new approaches to structure determination, especially in systems of potential technological importance such as fuel cells. Nuclear magnetic resonance (NMR) spectroscopy, which has a long history in solid-gas interfacial studies, is one technique which can, in principle, give substantial amounts of information about the solid-liquid interface, and previously some preliminary results were reported [2, 3], showing feasibility.

The strengths of NMR lie in the detailed types of information which can be derived. These include structural and motional information with regard to the adsorbate being studied, e.g., what is the adsorbate [4]? Is there any motion such as diffusion or internal motion? Furthermore, in the case of adsorbates on metals, information regarding the electronic environment can also be obtained from the Korringa relation [5], which in suitable situations can be related to densities of states at the Fermi level. Thus, the use of NMR to study adsorbates at the electrochemical solid-liquid interface represents a logical and systematic progression in the investigation of adsorbates on heterogeneous catalysts. The study of CO, adsorbed from the gas phase (solid/gas interface), on supported catalysts has received considerable attention previously [6, 7] and the methodologies in this area, at least from the NMR perspective, should be directly transferable to the solid-liquid interface. To illustrate this, such information has already been determined for electrochemically prepared sealed samples [8], where from Ti data it has been determined that the type of adsorbate derived from the electrodecomposition of methanol on polycrystalline (PC) platinum occupies a single adsorption site, regardless of the decomposition potential or surface coverage.

In this paper, we present recent work on the development of the electrochemical NMR cells for the study of electrode adsorbates as a function of applied potential. A number of novel design features permit the routine and reliable detection of adsorbed species, such as 13CO (ex MeOH) and CN" on fuel cell grade platinum black. Much higher sensitivity than previously reported is possible with our new instrumentation, which permits the recording of NMR frequency shifts as a function of applied potential. Although previous work from this laboratory demonstrated the feasibility of these techniques, it was limited in results and applicability in that the previous designs could only maintain the potential of the adsorbate during the NMR measurements, and sample preparation was accomplished in a separate cell suitable for the rigors of cyclic voltammetry. This restricted application to adsorbates which could withstand being transferred from a preparation cell into the electrochemical NMR cell. The older versions of the electrochemical NMR cell also suffered from poor sensitivity and required >105 transients for modest signal-to-noise ratio spectra. Finally, the old designs suffered from joints and Teflon stoppers that were positioned below the electrolyte level and often leaked, thus placing time constraints on spectral measurements, preventing the lengthy and elaborate NMR experiments required for detailed structure/dynamics characterization from being performed. The new cells permit routine operation as a function of applied potential, and form the basis for an even higher-field (600 MHz, 14) system under construction.

0 0

Post a comment