Sometimes a scientific experiment represents an idealized version of the real situation. This has long been recognized in catalysis research, where the study of catalytic site and mechanism in the idealized system differs substantially from the catalyst in action in a reactor1. This is due to limits in obtaining high-quality spectroscopic, structural or other data, which requires lower temperatures and pressures, and are measured over shorter timescales than occurs in reactor operation. That these gaps can lead to misleading conclusions regarding fundamental mechanisms was recognized when operando approaches were introduced in 2002 (ref. 2). In such approaches, the aim is to study catalysts under operating conditions. Given the complexity of energy materials, involving hierarchical length and time scales, operando studies of materials could lead to more information about materials function and performance at the reactor or device scale. In this issue of Nature Materials, we present three papers on operando studies of energy materials, with mechanistic insight into these systems in action.

The alternative text for this image may have been generated using AI.
Hexagonal Zn(002) platelet shows line-like stripping behaviour, where the heat map shows increased photon number produced from the fluorogenic reaction following Zn ion coordination bonding. Figure reproduced from the Article by Xianwen Mao and colleagues, Springer Nature Limited.

Electrochemical water oxidation through the oxygen evolution reaction (OER) can enable green hydrogen production. Iridium oxides (IrOx) are a model catalyst for the OER, exhibiting both high activity and stability. The elemental oxidation state of IrOx, and so information about chemical bonding and redox reactivity such as active states, are still under debate. In an Article by Reshma Rao and colleagues, time-resolved operando optical and X-ray spectroscopy are used to determine oxidation states. Density functional theory calculations show that at the high potentials relevant for the OER, iridium 5d states participate in this reaction, but oxygen 2p states also rise in energy to participate. Time-resolved measurements provide information in open-circuit conditions where OER reactivity decays, the optical fluorescence signals of Ir4+ and Ir5+ are stable, but the O−1 states rapidly decay with similar kinetics to OER decay, indicating that these surface oxygen ligand species are involved with O–O bond formation and oxygen release. In a related News & Views article by Alex Muñoz and Travis Jones, the role of iridium as a spectator during this reaction is noted and it is remarked that this is a chemical reaction driven by oxygen hole density. As they comment, this implies that the reaction mechanism is outside the computational hydrogen electrode model and Butler–Volmer kinetics, and ask can this mechanism be generalizable to other OER catalysts?

Another idealization in model studies is that materials are crystalline; however, amorphous systems are commonplace. This complicates structural analysis, especially in batteries where amorphous structures are used in high-performance systems. Among battery chemistries, Li–S batteries are of interest due to a combination of plentiful sulfur as a cathode and high energy density, but the complex redox processes involved in cycling result in sulfur species reaching the anode, causing battery failure. The usage of sulfurized polyacrylonitrile (SPAN) as the cathode suppresses this, but structural details of SPAN throughout cycling remain incomplete. In an Article by Enyuan Hu and colleagues, operando total scattering and pair distribution function analysis as well as X‑ray absorption spectroscopy are used to monitor SPAN formation and structure during synthesis and cycling. The first cycle causes some Li+ ions to replace edge protons, leading to irreversible capacity loss. Moreover, S–C bond formation prevents fusion of cyclic chains, instead enabling π–π stacking and, in turn, long sulfur chain reorganization into shorter chains that enables sulfur redox with high cyclability. In a related News & Views article by Sebastian Risse and Pouya Partovi-Azar the usage of disorder as an engineerable feature is postulated, while multiscale modelling to connect atomic-scale and mesoscale properties and structure is discussed.

Surface effects can impact electrochemical properties such as battery dynamics. However, operando imaging of spatially and temporally evolving electrochemical reactivities at particle–electrolyte interfaces is challenging. In an Article by Xianwen Mao and colleagues, an ion localization optical nanoscopy technique is demonstrated with 50 nm and 20 ms resolution. This is achieved by coupling in situ ion generation with fluorogenic ions, allowing single-ion dynamics imaging. When applied to operando stripping of Zn(002) anodes, the authors find substantial heterogeneity (pictured), for example line-like stripping, driven by a cooperative mechanism involving electronic and diffusive effects. These effects can be tailored to improve stripping uniformity.

Even with increased refinement and development of operando techniques, challenges remain such as improvements in spatial and temporal resolution, and extending these approaches to higher pressure or greater complexity environments. Operando measurements should not be treated as sole go-to techniques for materials investigation, building as they do on studies of more idealized systems. Certainly though, fundamental insight into the mechanisms and performance of materials in energy applications can be obtained, and so help hasten the urgently needed transition to greener energy sources.