CO Oxidation on Pt-Group Metals from Ultrahigh Vacuum to Near Atmospheric Pressures. 1. Rhodium

Feng Gao, Yun Cai, Kerrie Gath, Yuhao Wang, Mingshu Chen, Qiangsheng Guo, D. Wayne GoodmanThe Journal of Physical Chemistry C 2008 · DOI

TL;DR

This study investigates the CO oxidation reaction on Rhodium (Rh(111)) across a wide range of pressures, from ultrahigh vacuum (UHV) to near atmospheric conditions. The research clarifies the roles of surface oxygen and CO in influencing reaction rates, challenging previous notions of "superior oxide reactivity" and highlighting the high activity of a CO-uninhibited metallic surface.

What Problem Does This Paper Address?

The catalytic oxidation of carbon monoxide (CO) on platinum-group metals is a fundamental reaction in surface science and catalysis. While well-understood at low pressures, a comprehensive understanding at pressures relevant to industrial processes (near atmospheric) has been challenging due to the limitations of surface-sensitive techniques. Recent studies have proposed that surface oxides on metals like Rh, Pt, and Pd exhibit "superior reactivity" for CO oxidation, potentially operating via a Mars-Van Krevelen redox mechanism rather than the traditional Langmuir-Hinshelwood (LH) mechanism. This paper aims to resolve these discrepancies by systematically studying CO oxidation on a Rh(111) model catalyst under both low and high pressure conditions, using advanced surface-sensitive techniques to probe surface species and reaction kinetics. Specifically, it seeks to clarify the influence of surface oxygen coverage on reaction rates and to determine the most active surface phase under realistic reaction conditions.

Methods

The researchers employed a combination of sophisticated surface science techniques:

  • Polarization Modulation Infrared Reflection Absorption Spectroscopy (PM-IRAS): This technique allows for in-situ monitoring of surface species (specifically adsorbed CO) even at high pressures, providing information about their vibrational frequencies and, indirectly, their coverages.
  • X-ray Photoelectron Spectroscopy (XPS): Used to determine surface elemental composition and oxidation states, particularly for oxygen and Rh, to quantify surface oxygen coverage.
  • Batch Reactor Kinetics: High-pressure reactions were conducted in a batch reactor where the overall pressure change of reactants and products was monitored to determine reaction rates.
  • CO Titration Experiments: Used to probe the nature and reactivity of different oxygen species on the surface.

The experimental setup allowed for seamless transfer between UHV and high-pressure cells, enabling a direct comparison of reaction behavior across different pressure regimes.

Datasets

Not discussed in the paper. The study relies on experimental measurements rather than pre-existing datasets.

Key Insights and Ideas

  • CO Inhibition is Key: The study consistently shows that CO strongly inhibits the reaction, particularly at lower temperatures and pressures. This inhibition is a primary factor determining reaction rates.
  • Oxygen's Dual Role: Surface oxygen acts as a reactant, essential for CO oxidation. However, at higher coverages (approaching a full monolayer), it can also lead to deactivation of the Rh surface, likely through oxidation.
  • "Superior Oxide Reactivity" Questioned: The paper provides evidence against the notion that surface oxides are inherently more reactive than metallic surfaces for CO oxidation. Instead, the highest reactivity is observed on a CO-uninhibited metallic surface with a moderate oxygen coverage (around 0.5 ML).
  • Three Regimes at High Pressure: For stoichiometric or slightly oxidizing conditions (O2/CO ≤ 2), three distinct reaction regimes were identified: CO-desorption limited, mass-transfer limited, and a transient high-rate regime.
  • Pressure Gap is Not Evident for Rh: Contrary to some interpretations, the study suggests that a highly reactive CO-uninhibited metallic surface phase exists and is accessible at elevated pressures, indicating no "pressure gap" in reactivity for Rh.

Key Results

  • Low Pressure (≤ 2 × 10⁻⁴ Torr): CO conversion reached a maximum of ~20% of the CO flux within a narrow temperature range, occurring on an O-rich surface (oxygen coverage ≤ 0.5 ML). Higher oxygen coverages led to deactivation.
  • High Pressure (0.01-88 Torr):
    • O2/CO ≤ 2: The reaction exhibited three regimes: CO-inhibited (CO desorption limited), mass transfer limited, and a transient high-rate regime. Surface oxygen coverage did not exceed ~0.5 ML in these conditions.
  • O2/CO ≥ 5: Reactivity decreased dramatically at high temperatures due to oxidation. While a surface oxide phase showed some reactivity, a bulk oxide was found to be largely inactive.
  • CO Coverage at Maximum Rate (Rmax): Surface CO coverage at Rmax increased with increasing total pressure and O2/CO ratio.
  • Vibrational Frequency Shifts: PM-IRAS data revealed characteristic CO vibrational frequencies associated with different surface oxygen coverages, allowing for estimation of oxygen levels even at high pressures where XPS is not applicable. A feature at ~2085 cm⁻¹ was linked to oxygen coverages near 1 ML, and a feature at ~2130 cm⁻¹ to bulk Rh oxide.

Limitations and Future Work

  • Surface Oxide Characterization: While the study suggests the formation of surface oxides under certain conditions, definitively characterizing their exact structure (e.g., trilayer oxide) and distinguishing them from subsurface oxygen was challenging with the available techniques.
  • "Superior Oxide Reactivity" Nuance: The authors acknowledge that oxide phases can show higher rates than CO-inhibited metallic surfaces, but argue this enhancement is often modest and doesn't negate the superior reactivity of the CO-uninhibited metallic phase.
  • Future Work: The paper implicitly suggests further investigation into the precise nature of the surface oxide phases and their role in deactivation. Additionally, extending these studies to other Pt-group metals under similar high-pressure conditions would provide a broader comparative understanding. The authors also hint at parallel studies on Pd and Pt, which would be valuable for comparison.

Summary updated

Abstract

The CO oxidation reaction on Rh(111) was studied both at low pressures (≤2 × 10−4 Torr) under steady-state conditions and at high pressures (0.01−88 Torr) in a batch reactor at various gaseous reactant compositions. Surface CO and O coverages were determined using polarization modulation infrared reflection absorption spectroscopy (PM-IRAS) and X-ray photoelectron spectroscopy (XPS). CO titration experiments were also carried out on surfaces with known oxygen coverages. Both CO and O inhibition were evident at low pressures so that only within a relatively narrow temperature range were the reaction conditions optimized such that the CO conversion reached ∼20% of the CO flux to the surface. For high pressures and with stoichiometric or slightly oxidizing reactant ratios (O2/CO ≤ 2), the reaction fell into three regimes: (i) a CO-inhibited low temperature regime where the reaction rate was determined by CO desorption; (ii) a mass transfer limited regime at high temperatures; and (iii) a transient, high-rate regime lying between regimes (i) and (ii) where the reaction was not completely controlled by mass transfer limitation. For all reaction conditions investigated (when O2/CO ≤ 2), the surface oxygen coverage did not exceed ∼0.5 monolayer. With very oxidizing reactants (O2/CO ≥ 5), the reactivity of the Rh surface decreased dramatically at high temperatures due to oxidation. Furthermore, the so-called “superior oxide reactivity” for CO oxidation that has been proposed in several recent studies is not evident in this investigation.
BibTeX
@article{gao2008co,
  author = {Feng Gao and Yun Cai and Kerrie Gath and Yuhao Wang and Mingshu Chen and Qiangsheng Guo and D. Wayne Goodman},
  title = {{CO Oxidation on Pt-Group Metals from Ultrahigh Vacuum to Near Atmospheric Pressures. 1. Rhodium}},
  journal = {The Journal of Physical Chemistry C},
  year = {2008},
  doi = {10.1021/jp8077979},
}
Figure