Work-Function-Resolved Imaging of Relaxation Oscillations and Chemical Spillover in CO Oxidation over Platinum Surfaces

This study combines operando scanning electron microscopy with frequency-modulated Kelvin probe force microscopy to achieve the first work-function-resolved imaging of CO oxidation on platinum surfaces, revealing that chemical wave propagation is driven by relaxation-type oscillations characterized by rapid oxygen coverage onset and gradual CO-state relaxation.

The Gist

Imagine a platinum surface as a bustling dance floor where two types of dancers, Carbon Monoxide (CO) and Oxygen (O₂), are constantly trying to claim space. Their goal? To bump into each other, pair up, and turn into Carbon Dioxide (CO₂), which then floats away.

This isn't a chaotic free-for-all, though. Instead, the dancers organize themselves into waves. One moment, the floor is covered in CO dancers; the next, a wave of Oxygen dancers sweeps through, clearing the floor and reacting with the CO. This creates a mesmerizing, moving pattern of light and dark spots, similar to ripples in a pond or a snake slithering across the ground.

For decades, scientists have watched these waves, but they've been looking at them through a "foggy window." They could see the waves move, but they couldn't tell exactly what the dancers were doing or why the waves changed shape. They knew the "bright" spots were one thing and the "dark" spots were another, but the connection was guesswork.

The New Tool: The "Work-Function" Flashlight

In this study, the researchers from the Czech Republic and China built a super-powered microscope that acts like a two-in-one detective.

1. The Camera (SEM): This takes high-speed photos of the dance floor, showing the waves moving in black and white.
2. The Sensor (KPFM): This is the game-changer. Imagine a tiny, invisible finger (the AFM tip) hovering just above the dance floor. This finger doesn't just take a picture; it measures the electrical "mood" (called the work function) of the surface.

Think of the "work function" like the electric charge of the floor.

• When Oxygen is on the floor, the surface feels "heavy" with a high electrical charge.
• When CO is on the floor, the surface feels "lighter" with a low electrical charge.

By using this sensor while the camera films, the scientists could finally say with 100% certainty: "That dark spot? That's Oxygen. That bright spot? That's CO." No more guessing.

The Big Discovery: The "Relaxation" Wave

The most exciting part of the story is how these waves move. Scientists used to think the waves moved like a smooth, rolling sine wave (like a gentle ocean swell).

But this new "electrical finger" revealed something totally different. The waves are actually Relaxation Oscillations. Here is a simple analogy:

Imagine a slinky or a rubber band.

• The Fast Snap (Oxygen Arrival): When the Oxygen wave hits, it doesn't creep in slowly. It snaps onto the surface instantly, like a rubber band being released. The change is sharp, sudden, and violent.
• The Slow Creep (CO Return): After the reaction, the CO dancers don't rush back. They slowly, lazily creep back onto the floor, gradually taking over again. It's a long, drawn-out process.

The researchers found that this "Fast Snap, Slow Creep" pattern happens even at very low pressures, which was a surprise. It's like a heartbeat that has a very sharp "thump" and a very long "rest."

Why Does This Matter?

Think of a catalyst (like the platinum in a car's exhaust system) as a factory worker.

• If the worker is too busy (covered in CO), they can't do their job.
• If they are idle (covered in Oxygen), they aren't reacting.
• They need to switch between these states efficiently to keep the factory running.

This study shows that the "switching" isn't a smooth, predictable machine. It's a jumpy, uneven process that depends heavily on local conditions. Sometimes the switch happens fast; sometimes it drags on.

The Takeaway

Before this study, we were watching a movie of the reaction in black and white, guessing what the characters were thinking.
Now, thanks to this new "electrical flashlight" (KPFM), we are watching the movie in high-definition color with subtitles. We can see exactly how the chemical waves are built, how they jump and relax, and how the tiny details of the surface affect the whole reaction.

This helps engineers design better catalysts for cleaner cars and more efficient factories, because now they understand the rhythm of the reaction, not just the visual pattern.

Technical Summary

1. Problem Statement

Heterogeneous catalysis, specifically the oxidation of carbon monoxide (CO) on platinum (Pt) surfaces, exhibits complex spatio-temporal self-oscillations and chemical waves. While the macroscopic kinetics of these reactions are well-studied, the local electronic mechanisms driving the propagation of reaction fronts remain poorly understood under operando (working) conditions.

Key challenges addressed in this work include:

• The "Pressure Gap": Traditional surface science techniques (LEED, XPS, STM) often operate under ultra-high vacuum (UHV), failing to replicate industrially relevant pressures.
• Contrast Ambiguity in SEM: Scanning Electron Microscopy (SEM) can image reaction waves at high pressures, but the interpretation of secondary electron (SE) contrast is complex. It is difficult to unambiguously assign bright/dark regions to specific chemical states (CO-covered vs. O-covered) due to the interplay of work function, topography, and electric fields.
• Lack of Local Electronic Data: Existing techniques often provide spatially averaged data or lack the simultaneous high-resolution imaging and local electronic sensing required to resolve the internal structure of moving reaction fronts.

2. Methodology

The authors developed a novel correlative operando platform combining two distinct techniques within a single vacuum environment:

• Scanning Electron Microscopy (SEM): Used for wide-area, high-temporal-resolution imaging of surface topography and chemical contrast (secondary electron yield). Experiments were conducted on both Pt(110) single crystals and polycrystalline Pt wires at temperatures ranging from ~170°C to 216°C and pressures between $10^{-4}$ and $10^{-2}$ Pa.
• Frequency-Modulated Kelvin Probe Force Microscopy (FM-KPFM): Integrated directly into the SEM. A conductive AFM tip acts as a localized sensor to map the Contact Potential Difference (VCPD), which directly correlates to the local surface work function (WF).
  • Mode: The tip was held stationary while the reaction front propagated underneath it, allowing for time-resolved measurement of work function changes at a specific point.
  • Sensitivity: Capable of detecting work function variations with nanometer-scale spatial resolution.
• Computational Modeling: A spatially resolved Metropolis Monte Carlo (MMC) reaction-diffusion model was employed to simulate the kinetics. The model incorporated CO/O₂ adsorption, dissociation, surface diffusion, and the Langmuir-Hinshelwood reaction mechanism, running on GPU-accelerated hardware to match experimental conditions.

3. Key Contributions

• First Work-Function-Resolved Imaging of Reaction Fronts: The study provides the first simultaneous mapping of SE contrast and local work function variations during dynamic CO oxidation, enabling an unambiguous physical assignment of surface states.
• Identification of Relaxation Oscillations at Low Pressure: The authors identified relaxation-type oscillations (characterized by distinct fast and slow time scales) at low pressures ($10^{-2}$ Pa), a regime previously thought to exhibit only harmonic oscillations.
• Resolution of Chemical Spillover Asymmetry: The work reveals that the transition between chemical states is not symmetric. It demonstrates a rapid onset of oxygen coverage followed by a gradual, diffuse relaxation back to the CO-covered state.
• Local vs. Global Thresholds: The study challenges the mean-field assumption of a single global critical coverage threshold for reaction switching, showing instead that the transition depends on local kinetic environments and spatial heterogeneity.

4. Key Results

A. Correlative Imaging and State Assignment

• Unambiguous Assignment: By correlating SEM greyscale with FM-KPFM VCPD signals, the authors confirmed that:
  • High VCPD (High Work Function) corresponds to Oxygen-covered (O-Pt) regions.
  • Low VCPD (Low Work Function) corresponds to CO-covered (CO-Pt) regions.
• Contrast Limitations: While SEM shows a bimodal intensity distribution, it fails to resolve the internal structure of the wavefront. The transition appears sharp in SEM but is revealed by KPFM to have a complex, extended gradient.

B. Asymmetry and Relaxation Dynamics

• Temporal Asymmetry: The KPFM time traces reveal a pronounced asymmetry:
  • O-formation: A sharp, rapid transition (steep leading edge) when CO coverage drops below a critical local level.
  • CO-recovery: A slow, gradual decay (long tail) as the system relaxes back to the CO-covered state.
• Phase Portraits: Analysis of KPFM data using time-delay embedding produced triangular phase portraits, indicative of relaxation oscillations. This contrasts with the elliptical phase portraits typical of harmonic oscillations seen in previous Photoemission Electron Microscopy (PEEM) studies.
• Temperature Sensitivity: The triangular shape becomes more pronounced (approaching a right angle) with increasing temperature, highlighting the system's sensitivity to thermal conditions.

C. Simulation Validation

• The MMC simulations successfully reproduced the experimental spiral and planar wave patterns.
• Simulated time traces mirrored the experimental KPFM asymmetry, confirming that the observed dynamics arise from the nonlinear interplay of adsorption kinetics and lateral interactions, rather than artifacts.
• Simulations ruled out the formation of large-scale surface oxides, suggesting the reaction proceeds via chemisorbed atomic oxygen on metallic Pt (Langmuir-Hinshelwood mechanism).

D. Detection Limits

• Monte Carlo simulations of SE imaging demonstrated that while SEM has high spatial resolution, it lacks the sensitivity to detect the internal gradient structure of the reaction waves (nanometer-scale heterogeneities) due to signal averaging and low signal-to-noise ratios for work-function-induced contrast. KPFM is superior for resolving these internal electronic structures.

5. Significance

• Refining Catalytic Theory: The findings refine the classical mean-field description of oscillatory CO oxidation by emphasizing the importance of spatially resolved kinetics and local work function gradients.
• Mechanistic Insight: The observation of relaxation oscillations at low pressures suggests this behavior is more common than previously thought, likely masked by the lower resolution of traditional techniques.
• Methodological Advancement: The integration of SEM and KPFM establishes a powerful new paradigm for operando catalysis research, bridging the gap between wide-area imaging and localized electronic sensing. This allows researchers to distinguish active from inactive sites and understand the "hidden" internal structure of chemical waves.
• Implications for Catalyst Design: Understanding the local thresholds and the asymmetry of spillover processes provides critical insights for designing catalysts that can better manage metastable states and optimize reaction rates under dynamic conditions.