Discover how mechanical stress is transforming catalyst design and enabling unprecedented control over chemical reactions at the atomic level.
Imagine a world without modern fuels, plastics, or fertilizers—a world where essential chemical processes are too slow or too expensive to be practical. This would be our reality without catalysts, the unsung heroes of chemistry that accelerate reactions without being consumed. Catalysts are the silent partners in creating an estimated 90% of all manufactured materials around us, from the gasoline in our cars to the fertilizers that feed nations 7 .
For decades, scientists focused on two primary strategies to improve catalysts: changing their chemical composition or altering their physical structure.
Recent groundbreaking research has revealed a powerful new dimension to catalyst design: mechanical stress.
The concept is as simple as it is revolutionary—by subtly stretching or compressing catalyst materials at the atomic level, we can dramatically alter their ability to capture and transform molecules. This discovery of stress-controlled chemisorption (the chemical bonding of molecules to surfaces) is upending long-held beliefs in chemistry and opening unprecedented opportunities to design more efficient, selective, and cost-effective catalytic processes. From cleaning pollutants to producing green hydrogen, the implications for our energy and environmental future are profound.
At the heart of every catalytic reaction lies a critical step: chemisorption. Unlike physical adsorption (physisorption), where molecules are weakly attracted to surfaces like balloons sticking to a wall, chemisorption forms much stronger chemical bonds between the adsorbate (the molecule) and the adsorbent (the catalyst surface) 1 .
Think of the difference between a casual handshake and forming a business partnership—chemisorption creates a substantive chemical relationship that can alter the very nature of the molecules involved.
This process occurs when gas or vapor molecules move close enough to surface atoms that their electron clouds overlap, allowing them to share electrons and form genuine chemical bonds 1 . The energy involved is substantial—typically around 200 kJ/mol for chemisorption compared to a mere 20 kJ/mol for physisorption 1 .
Comparison of energy levels and bond strength between chemisorption and physisorption processes.
On any catalyst surface, certain locations called active sites possess special chemical properties that make them particularly effective at facilitating reactions. These aren't always obvious to the naked eye—they may be atomic steps, edges, defects, or low-coordination atoms that break the symmetry of an otherwise perfect crystal 2 .
In carbon-based catalysts, for instance, active sites arise from "defects such as vacancies that interrupt the sp2 carbon network, or topological distortions" 2 .
The performance of a catalyst depends critically on both the number and availability of these active sites to reactant molecules 5 . If sites are buried deep within pores too narrow for reactants to enter, they can't participate in the reaction. This is why catalyst designers strive to maximize accessible active sites, often by creating materials with high surface areas or strategic pore structures.
For decades, the prevailing theory about how strain affects catalysts centered exclusively on electronic effects. The widely accepted "d-band model" suggested that tensile strain (stretching) of late transition metals like nickel or platinum would narrow the d-band width of electrons, causing an upward shift that strengthened bonding with adsorbates 7 .
The rule was simple: tension strengthens binding, compression weakens it.
This understanding guided the design of core/shell nanoparticles, where lattice mismatch between core and shell materials intentionally strains the catalytic surface 7 . The electronic explanation seemed comprehensive—until researchers looked more closely at what happens at the steps and edges where so much catalysis actually occurs.
In 2015, a computational study published in Nature Communications sent ripples through the catalysis community. Researchers asked a simple but profound question: Does mechanical strain affect binding energy the same way everywhere on a catalyst surface? 7
Their approach was meticulous. Using molecular simulations, they studied the chemisorption energies of various molecules (CO, SH, OH, NH) on different metal surfaces under controlled strain conditions. They examined both flat (111) surfaces and stepped (211) surfaces under three types of stress: uniaxial stress perpendicular to steps, uniaxial stress parallel to steps, and biaxial stress across the entire surface 7 .
The researchers decomposed the binding energy changes into two components: an "electronic" contribution (from surface-adsorbate electronic interactions with fixed substrate ions) and a "mechanical" contribution (from subsequent relaxation of the substrate ions) 7 . This elegant separation allowed them to isolate effects that had previously been conflated.
The findings defied decades of established theory. While flat surfaces followed the expected pattern—tension strengthened binding—stepped surfaces displayed the opposite behavior: compression strengthened binding, and the effect was substantial 7 .
Even more remarkably, the response depended on the direction of applied stress. For CO on Cu(211), stress applied normal to the step lines produced one trend, while stress parallel to the steps produced another 7 . This directional dependence had never been observed before.
Most significantly, the researchers discovered that the mechanical energy change could be larger than and opposite in sign to the electronic energy change, leading to a complete reversal of the expected trend 7 . The mechanical effect wasn't just a minor correction—it could dominate the entire response.
| Surface Type | Strain Type | Effect on Binding |
|---|---|---|
| Flat (111) | Tensile | Increases |
| Flat (111) | Compressive | Decreases |
| Stepped (211) | Tensile | Decreases (opposite!) |
| Stepped (211) | Compressive | Increases (opposite!) |
| Catalyst | Strain | Activity |
|---|---|---|
| Nickel | Unstrained | Moderate |
| Nickel | 3% Tension | Peak |
| Cobalt | Unstrained | Peak |
| Ruthenium | Unstrained | Peak |
The unexpected behavior at steps has a physical origin. When adsorbates bind to step sites, they induce significant local reconstruction of the surface atoms. Under applied strain, the energy required for this reconstruction either adds to or subtracts from the overall energy change, depending on whether the strain is tensile or compressive 7 .
Think of it like this: binding to a step site already distorts the local atomic arrangement. If the surface is pre-compressed, this distortion comes "for free" mechanically, lowering the overall energy cost. If the surface is pre-stretched, the binding must work against the existing strain, increasing the energy cost.
This mechanical contribution had been overlooked because previous models assumed electronic effects dominated. But for many step-binding configurations, the mechanical energy is the primary driver, strong enough to overcome and reverse the electronic trend 7 .
The practical significance of this discovery becomes clear in reactions like methanation (converting synthesis gas to methane), which occurs primarily at step sites. The researchers computed that applying biaxial tension to nickel catalysts could shift their activity to the peak of the methanation "volcano plot"—making nickel as active as expensive ruthenium and cobalt catalysts 7 .
This demonstrates how mechanical strain can tune a catalyst to optimal performance by independently adjusting the binding strengths of different reaction intermediates—something composition changes cannot achieve as precisely.
Studying stress-controlled chemisorption requires sophisticated techniques that can probe both atomic structure and chemical activity. Researchers employ an array of specialized instruments and methods:
Measures binding strength by tracking desorption during heating. Identifies active sites and determines adsorption energy distribution.
Measures gas uptake at constant temperature. Quantifies number of active sites and surface energy distribution.
Direct visualization of catalyst changes during reaction. Reveals nanoscale structural transformations in real time.
Analyzes surface chemistry and oxidation states. Determines elemental composition and nature of active sites.
Computational modeling of atomic-scale interactions. Predicts binding energies and reveals mechanical/electronic contributions.
Flowing gas technique to measure active sites. Determines metal dispersion and active surface area.
These tools have revealed that the world of catalysts is far more dynamic than previously thought. For instance, recent studies using correlated operando microscopy and spectroscopy have shown that catalysts can maintain unexpected mixed phases during operation rather than transforming into a single "active state" 4 .
Copper-based pre-catalysts for nitrate reduction, for instance, can persist as a mixture of metal, oxide, and hydroxide throughout the reaction, with the composition depending on electric potential and chemical environment 4 .
Similarly, research on vinyl acetate production has revealed that palladium catalysts don't operate solely as solid surfaces but actually cycle between solid and soluble molecular forms in a "cyclic dance" driven by corrosion and electrodeposition processes . This blurs the traditional boundary between heterogeneous and homogeneous catalysis.
The principles of stress-controlled chemisorption extend beyond metal catalysts. Carbon-based catalysts used in thermo-catalytic decomposition (TCD) of natural gas demonstrate how structural disorder creates active sites. Unlike metals, carbon catalysts lack well-defined crystalline structures, and their active sites arise from "lattice defects, low-coordination atoms, edge sites, vacancies, and dislocations" 2 .
The more disordered the carbon structure, the more active it tends to be. Studies show TCD activity follows this order of structural order: amorphous > turbostratic > graphitic 2 . This is because disordered carbons contain smaller crystallites with more edge sites and "free valences" where reactions can occur.
Activity increases with structural disorder in carbon catalysts.
During TCD, the carbon deposit itself becomes the catalyst in an "autogenic" process, with the accumulating carbon continuously modifying the catalyst surface 2 . The long-term activity depends on how the evolving deposit maintains active site availability through self-renewal—a process potentially influenced by mechanical stress at the atomic scale.
A particularly promising area for stress engineering is high-entropy alloys (HEAs)—materials containing five or more elements in nearly equal proportions 6 . The inherent lattice distortion in these materials creates natural strain fields that can be precisely tuned by adjusting element combinations and proportions 6 .
The "cocktail effect" in HEAs—where the combination of elements produces unique properties not found in any constituent—makes them ideal platforms for stress-optimized catalysis 6 . However, their incredible compositional complexity presents a challenge: with millions of possible surface configurations, traditional experimental approaches struggle to identify optimal compositions.
This is where machine learning is becoming indispensable. ML models can establish "efficient, scalable mappings from composition, structure or site environment to HEA properties," including adsorption energies 8 . By training on smaller DFT datasets, these models can predict adsorption energies across the vast configuration space, dramatically accelerating the discovery of strain-optimized HEA catalysts 8 .
The discovery that mechanical stress can control chemisorption represents a paradigm shift in catalysis. What was once viewed through a purely electronic lens now embraces mechanical effects that can dominate, and even reverse, established trends. This more complete understanding opens exciting possibilities for designing catalysts with atomic precision.
More efficient processes across chemical manufacturing
Cleaner energy and pollution control technologies
Atomic-scale control over catalytic properties
The implications extend across the chemical industry—from more efficient fertilizer production that reduces energy consumption to improved fuel cells and environmental cleanup technologies. As researchers increasingly recognize that catalysts are dynamic, not static, and that surfaces can undergo complex transformations during operation, our ability to design better catalysts grows exponentially.
The most promising realization is that mechanical strain provides an independent design parameter that complements traditional composition and structure approaches. By thoughtfully combining all three—composition, structure, and strain—we enter a new era of catalyst design where atomic-scale engineering can create materials with precisely tailored catalytic properties.
As we continue to unravel the complex interplay between stress and surface chemistry, we move closer to a future where chemical processes are more efficient, more selective, and more sustainable—all by appreciating the profound influence of the tiny mechanical forces that shape the molecular world.
References will be added here manually as needed.