Catalysis at Surfaces
Over 90% of industrial chemical processes rely on heterogeneous catalysis — reactions accelerated by solid surfaces. From ammonia synthesis feeding billions to catalytic converters cleaning exhaust, these processes depend on the precise interaction between reactant molecules and catalyst surfaces. The key insight, formulated by Paul Sabatier in 1911, is that the best catalyst provides an intermediate level of interaction: strong enough to break bonds in reactants, weak enough to release the products.
The Volcano Plot
When catalytic activity is plotted against the binding energy of a key intermediate, a volcano-shaped curve emerges. On the ascending (left) side, stronger binding increases surface coverage and lowers activation barriers. On the descending (right) side, binding is so strong that products cannot desorb, poisoning the surface. The peak represents the Sabatier optimum — and remarkably, the same metals (Pt, Pd, Ru) sit near the peak for many different reactions.
The BEP Relation
The Bronsted-Evans-Polanyi (BEP) relation provides the theoretical backbone of volcano plots. It states that activation energy decreases linearly with increasing binding energy: stronger binding makes it easier to break reactant bonds. Combined with the Langmuir adsorption model, this creates the fundamental tradeoff: increasing binding energy simultaneously lowers the activation barrier (faster reaction) but increases surface coverage (fewer available sites), producing the volcano shape.
Designing Better Catalysts
Modern computational catalysis uses density functional theory (DFT) to calculate binding energies and screen thousands of candidate materials. Descriptor-based design identifies the optimal binding energy from volcano plots, then searches for alloys and nanostructures that achieve it. This approach has led to discoveries of non-precious metal catalysts for fuel cells, electrochemical CO₂ reduction, and nitrogen fixation — potentially replacing expensive platinum group metals with abundant alternatives.