In chemistry an important quantity is the rate of a reaction, the number of product molecules formed per second, which averages over myriad molecular rearrangements. The motions of the atoms themselves are not accessible to classical analytical methods. A group of scientists at the Fritz Haber Institute in Berlin achieved to view the individual particles during a chemical reaction. This work, published in the Dec. 12, 1997 issue of Science, directly relates atomic processes with quantities measurable by ordinary chemistry.
The authors used a scanning tunneling microscope that allows atomic resolution on solid surfaces to investigate the reaction between oxygen atoms and carbon monoxide molecules to carbon dioxide on the surface of a platinum crystal. It is believed that this reaction plays also an important role in the removal of CO from exhaust gases by Pt catalysts.
This process belongs to a class of chemical reactions known as heterogeneous catalysis in which reaction rates are dramatically enhanced at surfaces of particles, often of certain metals, which are added to the reaction mixture. Because of their enormous role in chemical industry these reactions have been studied for several decades in an attempt to understand the underlying mechanisms, the sequence of steps that finally lead to the desired products. It turned out that even apparently simple reactions like the Pt catalyzed CO oxidation are quite complex and proceed through several intermediate steps, many of which are still unknown, to the final product. In all cases the central step is a reaction between atoms or molecules bound to the surface of the catalyst. However, there have been indirect indications that this central process might also be more complicated than assumed in the simplified models used to describe industrial catalytic reactions.
The scanning tunneling microscopy data show that this is in fact the case: The surface oxygen atoms and CO molecules which are resolved by the scanning tunneling microscope separate into islands, patches containing only one of the two reactive species, so that the reaction can take place only at the borders of these islands.
By counting the number of oxygen atoms that are removed from given areas of the surface by the reaction with CO it was possible to evaluate reaction rates that are not subject to the averaging problems encountered in ordinary chemical experiments. The authors demonstrate that the rates thus determined are fully consistent with previously measured values when the effect of the island separation is properly taken into account.
Journal
Science