In Situ Gas Phase Electron Microscopy Resolves Catalyst Sintering on Oxide Supports

Direct real-time imaging using in situ gas phase transmission electron microscopy has allowed researchers to directly track the sintering behavior of metal catalyst nanoparticles on oxide supports under high-temperature reaction conditions. Metal nanoparticle catalysts are critical components in industrial catalytic conversion and vehicle catalytic converters, where they facilitate chemical reactions to reduce harmful gaseous pollutants such as carbon monoxide, nitrogen oxides, and hydrocarbons emanating from vehicle exhaust, aircraft engines, and industrial facilities. In operating vehicles, catalytic converters generally require annealing temperatures ranging from 280 degrees Celsius to 650 degrees Celsius under atmospheric pressure conditions to maintain proper chemical conversion. Reducing the dimensions of metal nanoparticles increases the proportion of exposed surface active sites, which in turn enhances overall catalytic efficiency. However, when metallic catalyst nanoparticles encounter high thermal thresholds, specifically temperatures recognized as the Tammann temperature, their constituent metal atoms gain sufficient mobility to diffuse across the oxide support surface, detach from interatomic bonds, and merge with adjacent particles in the degrading process known as sintering.

Thermal sintering of supported metal nanoparticles proceeds primarily through two distinct physical mechanisms that coalesce particles and decrease active surface area, thereby impairing catalytic performance. Under Ostwald ripening, smaller metal nanoparticles gradually disappear as their mobile constituent atoms detach and migrate across the support or through the gas phase to deposit onto larger, surviving particles. In contrast, under particle migration and coalescence, entire metal nanoparticles break their bonding interactions with the underlying support, migrate across the surface, and merge upon contact with neighboring particles to form larger aggregates. To suppress both sintering pathways, catalyst formulations routinely disperse metal nanoparticles onto oxide supports such as titanium dioxide, zirconium dioxide, cerium dioxide, or manganese dioxide. These oxide materials exhibit lower sintering tendencies than bulk metal nanoparticles while offering high corrosion resistance, robust thermal stability, and durable mechanical properties. Modifying the surface structure of these oxide supports strengthens the metal-support interaction, anchoring metal nanoparticles through surface bonding and substantially improving their thermal sintering resistance.

Historically, verifying the theoretical models of Ostwald ripening and particle migration and coalescence remained challenging due to the scarcity of high spatial resolution imaging capable of recording immediate nanoparticle dynamics under simulated operating conditions. While X-ray absorption spectroscopy effectively characterizes bonding between catalyst nanoparticles and oxide supports to correlate metal-support interaction enhancements with sintering resistance, it does not provide dynamic, real-time spatial imaging of individual migrating particles. Conventional transmission electron microscopy determines particle size distributions across multiple static images taken after reaction stages, yet cannot unambiguously distinguish intermediate sintering mechanisms. Scanning tunneling microscopy enables direct observations of sintering progression during temperature shifts, but its operational principles restrict analysis strictly to conductive materials. To bridge the divide between theoretical sintering models and physical observation, researchers developed in situ gas phase transmission electron microscopy, which allows direct monitoring of catalyst particle size, particle count, and interparticle distance at temperatures exceeding 1000 degrees Celsius under atmospheric pressure gas environments.

Applying in situ gas phase transmission electron microscopy requires specialized instrumentation that reconciles high-pressure gas environments with the high-vacuum requirements of the electron column. Standard transmission electron microscopes operate in vacuum to preserve the electron beam path, whereas catalyst reactions occur in gaseous atmospheres. Researchers deploy two primary instrument designs to overcome this barrier: the environmental transmission electron microscope and the windowed gas cell holder. Environmental transmission electron microscopy introduces gases directly into the chamber between the objective lens polepieces, relying on differential pumping through small apertures to contain gas leakage while utilizing conventional sample holders and heating chips. While this design avoids beam-scattering window membranes and achieves high spatial resolution, its maximum gas pressure is generally restricted to approximately 100 Torr. Because elevated gas pressures in large column volumes increase electron scattering and degrade spatial resolution, environmental transmission electron microscopy cannot fully simulate full atmospheric production environments.

The windowed gas cell configuration provides an alternative approach by sealing the reaction atmosphere within a microelectromechanical systems holder equipped with two electron-transparent silicon nitride windows. The catalyst specimen is placed between the two aligned silicon nitride chips, isolating circulating gases from the high-vacuum column without requiring supplementary column vacuum pumps. This closed design accommodates gas pressures near ambient atmospheric conditions, replicating realistic industrial reaction environments to quantify sintering resistance accurately. However, the transmission of the electron beam through both silicon nitride windows causes beam scattering that decreases image resolution. Furthermore, gas flow within the restricted cell can induce window bulging, widening the internal gap and forcing the beam to collide with additional gas molecules, which exacerbates scattering and reduces spatial resolution. Specimen preparation also requires meticulous precision to mount catalysts onto delicate silicon nitride surfaces and achieve exact optical alignment between the microelectromechanical windows.

In situ gas phase electron microscopy has proven instrumental in evaluating how surface defects on oxide supports anchor metallic nanoparticles against thermal degradation. Theoretical density functional theory calculations established that surface steps and vacancies introduce negative interaction energies, creating preferential binding sites that hinder nanoparticle diffusion across the oxide. Real-time environmental transmission electron microscopy tracking of gold nanoparticles supported on titanium dioxide at 700 degrees Celsius under 0.05 Pascal of oxygen demonstrated significant sintering through concurrent Ostwald ripening and particle migration and coalescence, as small particles vanished while adjacent particles coalesced over a 60-minute duration. Conversely, annealing identical gold-titanium dioxide catalysts under a carbon monoxide atmosphere at 700 degrees Celsius and 0.05 Pascal produced no measurable alterations in nanoparticle size or distribution. The injected carbon monoxide induced inhomogeneous void and step defects across the titanium dioxide surface, and the resulting vacancies anchored the gold atoms, strengthening metal-support interactions and preventing sintering.

Introducing rare earth dopants into oxide supports represents another surface modification strategy confirmed through in situ gas phase transmission electron microscopy. Spectroscopic characterizations including extended X-ray absorption fine structure and temperature-programmed reduction showed that incorporating neodymium into lanthanum-zirconium oxide supports generates strong rhodium-oxygen-neodymium bonds, preserving nanoparticle dispersion throughout high-temperature redox aging tests. In related in situ scanning transmission electron microscopy investigations utilizing a windowed gas cell, rhodium nanoparticles on pure zirconium dioxide exhibited noticeable sintering beginning at 900 degrees Celsius under nitrogen and oxygen atmospheres, with particle diameters expanding proportionally as temperatures reached 950 degrees Celsius. In contrast, rhodium nanoparticles supported on yttrium-doped zirconium dioxide maintained stable diameters under identical heating parameters. Pure zirconium dioxide acts as an electrical insulator, whereas yttrium doping imparts ionic conductivity that induces back-spillover of oxygen species toward the metal nanoparticles at high temperatures, increasing oxygen coordination and strengthening interfacial bonding.

The specific crystallographic facet exposed on the oxide support surface directly dictates diffusion energy barriers, metal-support bonding strength, and resulting catalytic activity. Density functional theory calculations indicate that crystal facets possessing higher densities of dangling bonds exhibit elevated surface energy, producing steeper diffusion barriers that restrict nanoparticle mobility. Environmental transmission electron microscopy tracking gold nanoparticles on titanium dioxide under oxygen at 500 degrees Celsius confirmed that particles deposited on the (101) facet sintered via Ostwald ripening and coalescence, whereas particles situated on the (001) facet resisted sintering completely. Similarly, in situ gas phase transmission electron microscopy of palladium nanoparticles supported on zinc oxide demonstrated that annealing in air from 200 degrees Celsius to 600 degrees Celsius caused severe sintering on the (100) facets of zinc oxide nanorods via particle migration and Ostwald ripening. In contrast, palladium nanoparticles supported on the (001) facets of zinc oxide nanosheets remained stable without sintering, preserving superior carbon monoxide oxidation rates across light-off curves evaluated after 200, 400, and 600 degrees Celsius calcination steps.

Encapsulation techniques that wrap nanoparticles in thin oxide or organic protective shells provide another validated pathway to reinforce metal-support interactions and stabilize active sites. In classical strong metal-support interactions, reducible oxide supports such as titanium dioxide generate thin overlayers that migrate over supported noble metal nanoparticles during high-temperature reductive treatments. In situ gas phase microscopy observations revealed that palladium nanoparticles on titanium dioxide heated to 250 degrees Celsius under a reducing 5 percent hydrogen in argon atmosphere became surrounded by an amorphous titanium suboxide layer that crystallizes at 500 degrees Celsius into an epitaxial overlayer along the palladium (111) plane. Although this encapsulation mechanically anchors nanoparticles against detachment at high temperatures, crystalline overlayers can block gas molecule access to active sites, thereby depressing carbon monoxide oxidation rates. In situ gas phase transmission electron microscopy observations also confirmed that the formation of such protective encapsulation layers depends on the underlying support facet, occurring on titanium dioxide (101) and (100) facets under oxidizing conditions at 550 degrees Celsius but failing to form on (001) facets.

To prevent active site blockage associated with permanent oxide overlayers, sacrificial organic coatings such as polydopamine have been developed and evaluated using windowed gas cell transmission electron microscopy. Heating polydopamine-coated titanium dioxide supports bearing palladium nanoparticles to 900 degrees Celsius for 90 seconds under argon inside the windowed gas cell converted the organic layer into a nitrogen-doped carbon shell while simultaneously generating surface oxygen vacancies. Tracking contrast reductions in real-time transmission electron microscopy demonstrated that nitrogen defects within the carbon shell trapped diffusing palladium atoms following Ostwald ripening detachment, dispersing them into isolated single atoms. Subsequent calcination at 500 degrees Celsius in air completely removed the sacrificial nitrogen-doped carbon shell, transferring the captured palladium single atoms directly into the oxygen vacancies of the underlying titanium dioxide support. By combining real-time gas phase transmission electron microscopy with tailored support modifications, researchers can validate the operative mechanisms of Ostwald ripening and particle migration while designing resilient catalyst architectures that sustain catalytic efficiency.

Subject of Research: In situ gas phase transmission electron microscopy of catalyst nanoparticle sintering resistance on oxide supports

Article Title: In situ observation of catalyst nanoparticle sintering resistance on oxide supports via gas phase transmission electron microscopy

Article References: Kim, W., Kim, K., Kim, J., & Lee, Z. (2024). In situ observation of catalyst nanoparticle sintering resistance on oxide supports via gas phase transmission electron microscopy. Applied Microscopy, 54(1), Article 7. https://doi.org/10.1186/s42649-024-00100-4

Image Credits: AI Generated

DOI: 10.1186/s42649-024-00100-4

Keywords: in situ TEM, environmental TEM, windowed gas cell, catalyst sintering, sintering resistance, Ostwald ripening, particle migration and coalescence, metal-support interaction, oxide supports, surface defects, rare earth doping, crystal facets, nanoparticle encapsulation, strong metal-support interaction, titanium dioxide, zirconium dioxide, zinc oxide, carbon monoxide oxidation