Spatiotemporal concentration patterns in a surface reaction: Propagating and standing waves, rotating spirals, and turbulence
In chemistry, two dimensions are often better than three, since surface-bound reactions can be probed in greater detail than those in a liquid solution. Gerhard Ertl was awarded the chemistry Nobel Prize last week for his many contributions to the field of surface chemistry. A professor emeritus at the Fritz Haber Institute of the Max Planck Society in Berlin, Ertl devoted years to understanding many surface-mediated reactions important in atmospheric science and industry, including the reaction that cleans up the fumes in a car’s exhaust. Three important papers in those studies appeared in Physical Review Letters, beginning in 1985.
“Ertl’s work is very fundamental,” says Harm Rotermund of Dalhousie University in Halifax, Canada, who worked with him on several projects. Rotermund says that Ertl’s curiosity led him to study complicated problems, applying all the tools available and developing new ones when necessary. His research touched upon several real-world applications, including nitrogen fixation on iron grains for making fertilizer, ozone-depleting reactions on ice crystals, and the transformation of toxic carbon monoxide (CO) into carbon dioxide by a car’s catalytic converter.
Ertl began working on this last reaction, so-called CO oxidation, in the early 1980s. Researchers observed a decade earlier that in the reaction of CO with oxygen on a platinum surface, the rate of reaction isn’t constant but oscillates up and down with time. Although oscillating reactions were known in liquids, this was the first known example on a surface. “No one knew what was going on,” says Ronald Imbihl from the University of Hannover in Germany, so Ertl simplified the problem by placing the reactants in ultra-high vacuum with single platinum crystals. Using low energy electron diffraction (LEED), he and his colleagues, including Imbihl, discovered that the platinum rearranges its surface structure to accommodate CO molecules that stick to it. This surface reconstruction increases the amount of oxygen that can be trapped on the surface, which increases the reaction rate. As more CO is converted to CO coverage eventually drops, and the surface reverts back to its original structure, when the cycle starts over.
Ertl’s explanation of the oscillations, published in 1985, suggested that certain regions on the surface could be covered mostly with CO, while others mostly with oxygen. To confirm this, Ertl, along with Rotermund and others, redesigned a photoemission electron microscope (PEEM) to take pictures of the platinum surface with sub-micron resolution. The image came from electrons that had been stripped from the surface by an ultraviolet light. Bright and dark spots in the image corresponded, respectively, to CO and oxygen rich areas on the surface, because stripped electrons from oxygen-covered areas had less energy. “I remember how stunned we were when we got back the first images,” Rotermund says. The dark and light regions were organized into traveling waves, rotating spirals, and even chaotic patterns, as they described in their 1990 paper.
The many patterns illustrated characteristics of other so-called non-linear systems, like the weather, galaxy formation, and heart arrhythmias, which in many cases are considered unpredictable. In subsequent work, Ertl made the platinum surface his work-bench for probing and even controlling the non-linearity of CO oxidation, as exemplified in his 2004 paper. This kind of detailed study, not possible in three-dimensional solutions, is just one example of Ertl’s ability to build well-defined, stripped-down models to explain more complex reactions, Imbihl says. “His idea was that nature is simple.”
Michael Schirber is a freelance science writer in Lyon, France.