Many chemical processes that sustain modern society depend on molecules binding to metal surfaces. The production of synthetic fertilizer, for example, relies on heterogeneous catalysis, in which gases react on the surface of a solid catalyst. Yet, up until now an important question has remained unresolved: how much of the metal surface actually participates in determining the strength of a chemical bond on the surface?
A research team from the Max Planck Institute for Multidisciplinary Sciences (MPI-NAT) and the Center for Quantum Nanoscience (QNS) within the Institute for Basic Science (IBS) has now shown that bonding at metal surfaces extends beyond the atoms immediately surrounding the molecule.
Using precisely structured platinum surfaces, the researchers found that the strength of a molecule's bond at an atomic step can change when another step is moved as far as approximately 10 nanometers, away. The effect depended on the molecule and altered adsorption energies by up to 0.2 electron volts. The findings were published in Science.
Chemical bonding with a surprisingly long reach
Chemical bonding is often pictured as a local interaction. A molecule lands on a surface and forms a bond with one or a few nearby atoms. For many purposes, this local picture works well.
Bonding to a metal, however, is fundamentally different from bonding between isolated atoms or molecules. The electrons in a metal occupy electronic states that extend throughout the solid. When a localized molecular orbital interacts with these delocalized metal states, the electronic structure involved in the bond can extend far beyond the adsorption site.
An analogy is the difference between pressing a finger into sand and pressing it against a stretched drum skin. In sand, the disturbance remains largely confined to the grains being touched. On a drum skin, the deformation spreads over a much larger area because the surface is mechanically connected. In a metal, it is the delocalized electronic structure that provides this connection.
Long-range electronic disturbances around atoms and molecules on metal surfaces have been observed before, particularly in low-temperature studies. However, such effects were generally not considered to make a substantial contribution to chemical bond strengths under conditions relevant to catalysis. The present work directly shows that the strength of a molecule-metal bond itself can remain sensitive to structural variations many atomic spacings away.
Measuring the bond's "footprint"
To measure this effect, the researchers used platinum surfaces containing regularly spaced atomic steps, which are sites where molecules often bind particularly strongly. By preparing surfaces with different distances between neighboring steps, the team effectively used the platinum lattice as an atomic-scale ruler. They then measured the adsorption energies of ammonia (NH₃), carbon monoxide (CO), and nitric oxide (NO) using Velocity-Resolved Kinetics, a technique that determines how rapidly molecules leave a surface and used this information to extract their binding energies.
The results showed that changing the distance between neighboring steps changed the strength of the bond at a given step, even when the neighboring step was several nanometers away. The characteristic distance over which the effect persisted was molecule-dependent, reaching approximately 7 nanometers for NH₃, 6 nanometers for CO, and 4 nanometers for NO.
To explain why the effect differed from molecule to molecule, the researchers developed a Frontier Orbital-Band Model that combines the molecular-orbital view of chemistry with the electronic-band view of solid-state physics. The model showed that different molecules couple to different electronic states in the metal, causing each molecule to respond differently to structural changes elsewhere on the surface. The team also compared platinum with palladium and found that the long-range effect was much weaker in palladium, supporting the idea that it depends on how strongly electronic states extend through the metal.
Why it matters for catalysts
The findings may be particularly important for nanoscale catalysts. Catalytically active platinum nanoparticles often measure only a few nanometers across — comparable to the length scale over which the researchers found that bond strengths remain sensitive to surface structure. On a particle this small, an adsorbed molecule may therefore be influenced by structural features across a substantial fraction of the entire nanoparticle rather than only by the atoms immediately surrounding its binding site.
The work suggests that models of catalytic activity based only on the local arrangement of atoms around an adsorption site may provide an incomplete description for metals with strongly delocalized electronic bands. A more accurate picture must also consider how the surrounding solid modifies the electronic states involved in the bond.
The result grew out of a collaboration between MPI-NAT, where the high-precision measurement methods were developed, and QNS, which brought a surface-physics perspective to the problem. Dmitriy BORODIN, who co-developed the measurement methods during his doctoral research at MPI-NAT before joining QNS, helped bring the two perspectives together. Florian NITZ, PhD student at the University of Göttingen and lead author of the study, led the development of the Frontier Orbital-Band Model that explained the experimental findings. Together, these contributions led to a unified picture of how localized molecular orbitals interact with delocalized electronic states in a metal.
Rather than treating a catalytic bond as an interaction confined to a molecule and a few neighboring atoms, the study shows that the wider electronic structure of the metal can influence bond strength over distances of several nanometers. This broader view could help improve theoretical models used to understand and design nanoscale catalysts.
