Nonidealities in CO2 Electroreduction Mechanisms Revealed by Automation-Assisted Kinetic Analysis

Joy S. Zeng, Vineet Padia, Grace Y. Chen, Joseph Maalouf, Aditya Limaye, Alexander H. Liu, Michael A. Yusov, Ian W. Hunter, Karthish ManthiramACS 2024 · DOI

TL;DR

This paper introduces a novel robotic system for automating electrochemical kinetic analysis, enabling the collection of extensive datasets under varied conditions. This system reveals complex, non-ideal reaction mechanisms in CO₂ electroreduction to CO at metal tetrapyrrole catalysts, which are often missed by traditional analysis methods.

What Problem Does This Paper Address?

Understanding the intricate mechanisms of electrocatalytic reactions, particularly for CO₂ reduction, is crucial for developing efficient catalysts for energy conversion. Traditional methods for analyzing reaction kinetics, like Tafel plots and reactant order dependence, rely on simplifying assumptions that often fail when dealing with complex phenomena such as surface coverage effects or mixed reaction controls. These simplifications can lead to incomplete or misleading interpretations of how catalysts actually work. The challenge lies in collecting enough high-quality data across a wide range of experimental conditions to perform more robust, "cohesive" kinetic analyses, which are common in thermocatalysis but less so in electrocatalysis due to experimental hurdles.

Methods

The core innovation here is a robotic platform designed for automated, sequential electrochemical experiments. This system can test up to 10 different electrochemical cells, each with the potential for a unique electrode, electrolyte, gas composition, and applied voltage. Key features include:

  • Automated Cell Handling: A robotic arm moves a cell body to sequentially engage with different cell pans, allowing for automated electrolyte filling, gas purging, and voltage application.
  • Online Product Quantification: A gas chromatograph is integrated to continuously monitor CO and H₂ production, providing real-time rate data.
  • Versatile Experimental Conditions: The system accommodates variations in electrode material, electrolyte composition, and gas-phase reactant concentrations, which is essential for detailed kinetic analysis.
  • Cohesive Kinetic Analysis: The extensive data collected is then analyzed using quantitative model fitting across a broad range of conditions, moving beyond simple linearizations.

Datasets

Not discussed in the paper. The paper focuses on generating new experimental data using their automated system.

Key Insights and Ideas

  • Automation is Key to Unlocking Complex Kinetics: The developed robotic system is not just about speed; it's about enabling the collection of the right kind of data (extensive, varied conditions) needed for advanced kinetic analysis that traditional methods can't handle.
  • "Nonidealities" are Prevalent: The study reveals that complex, non-linear, and potential-dependent kinetic behaviors are common in CO₂ electroreduction to CO using immobilized metal tetrapyrroles, challenging the assumption of simple reaction pathways.
  • Beyond Tafel Slopes: The observed kinetic trends (e.g., changing reactant orders with potential) cannot be explained by standard Tafel analysis, highlighting the limitations of this common technique for complex systems.
  • Bicarbonate and CO₂ Interactions are Complex: The way bicarbonate and CO₂ interact with the catalyst surface and influence reaction rates changes significantly with applied potential, suggesting dynamic surface chemistry.
  • Mechanistic Models Need to Account for Dynamic Effects: Phenomena like electrolyte poisoning and potential-dependent mixed control are crucial for explaining the observed kinetic data, indicating that catalyst behavior is not static.

Key Results

  • The automated system successfully collected kinetic data for three metal tetrapyrrole catalysts: cobalt phthalocyanine (CoPc), cobalt tetraphenylporphyrin (CoTPP), and iron phthalocyanine (FePc).
  • For CoPc and CoTPP, the observed bicarbonate and CO₂ order dependencies showed complex, voltage-dependent behavior, suggesting similar underlying mechanisms.
  • FePc exhibited distinct kinetic behaviors, particularly in its bicarbonate and CO₂ dependencies, suggesting a different most abundant reactive intermediate (MARI) compared to the cobalt catalysts.
  • The study provides qualitative mechanistic models that explain these complex trends, involving voltage-dependent poisoning and mixed control between different reaction pathways. For instance, FePc's mechanism appears to involve CO desorption as a rate-limiting step at lower potentials, leading to attenuated CO₂ order.

Limitations and Future Work

  • Qualitative vs. Quantitative Models: The authors acknowledge that their proposed mechanistic models are primarily qualitative and do not quantitatively fit all experimental data with statistical rigor. Future work could focus on developing more precise quantitative models.
  • Undivided Cell: The use of an undivided cell led to some loss of Faradaic efficiency (around 60% closure) due to potential parasitic oxygen reduction reactions. Future iterations could incorporate a membrane to separate anode and cathode.
  • Flow Configuration: While the study used bubbling for CO₂ delivery, the authors note that flow configuration can influence measured kinetics, and future work might explore different gas flow strategies.
  • Further Automation Development: The authors suggest extending the robotic concept to include membrane separation, alternative electrolyte/gas flow configurations, and automated mechanism enumeration and experimental design to further standardize and accelerate research.

Summary updated

Abstract

In electrocatalysis, mechanistic analysis of reaction rate data often relies on the linearization of relatively simple rate equations; this is the basis for typical Tafel and reactant order dependence analyses. However, for more complex reaction phenomena, such as surface coverage effects or mixed control, these common linearization strategies will yield incomplete or uninterpretable results. Cohesive kinetic analysis, which is often used in thermocatalysis and involves quantitative model fitting for data collected over a wide range of reaction conditions, requires more data but also provides a more robust strategy for interrogating reaction mechanisms. In this work, we report a robotic system that improves the experimental workflow for collecting electrochemical rate data by automating sequential testing of up to 10 electrochemical cells, where each cell can have a different electrode, electrolyte, gas-phase reactant composition, and applied voltage. We used this system to investigate the mechanism of carbon dioxide electroreduction to carbon monoxide at several immobilized metal tetrapyrroles. Specifically, at cobalt phthalocyanine (CoPc), cobalt tetraphenylporphyrin (CoTPP), and iron phthalocyanine (FePc), we see signatures of complex reaction mechanisms, where observed bicarbonate and CO<sub>2</sub> order dependences change with applied potential. We illustrate how phenomena such as electrolyte poisoning and potential-dependent degrees of rate control can explain the observed kinetic behaviors. Our mechanistic analysis suggests that CoPc and CoTPP share a similar reaction mechanism, akin to one previously proposed, whereas the mechanism for FePc likely involves a species later in the catalytic cycle as the most abundant reactive intermediate. Our study illustrates that complex reaction mechanisms that are not amenable to common Tafel and order dependence analyses may be quite prevalent across this class of immobilized metal tetrapyrrole electrocatalysts.
BibTeX
@inproceedings{zeng2024nonidealities,
  author = {Joy S. Zeng and Vineet Padia and Grace Y. Chen and Joseph Maalouf and Aditya Limaye and Alexander H. Liu and Michael A. Yusov and Ian W. Hunter and Karthish Manthiram},
  title = {{Nonidealities in CO<sub>2</sub> Electroreduction Mechanisms Revealed by Automation-Assisted Kinetic Analysis}},
  booktitle = {ACS},
  year = {2024},
  doi = {10.1021/acscentsci.3c01295},
}
Figure