Deciphering the Structure of Single Active Sites Under In Situ and Operando Conditions Using X-ray Absorption Spectroscopy

CatLab Lectures 2024/25

Date: Feb 21, 2025
Time: 10:30 AM - 12:00 PM (Local Time Germany)
Speaker: Dr. Andrea Martini
Fritz-Haber-Institute der Max-Planck-Gesellschaft
Location: Building M, Richard-Willstätter-Haus, Faradayweg 10, 14195 Berlin
Room: Seminar Room
Host: HZB and FHI
Contact: trunschk@fhi-berlin.mpg.de

Deciphering the Structure of Single Active Sites Under In Situ and Operando Conditions Using X-ray Absorption Spectroscopy

Understanding the structure of catalysts under relevant reaction conditions is crucial, as their active sites often undergo dynamic changes. These transformations can significantly influence the catalytic performance, emphasizing the importance of employing operando techniques that enable real-time monitoring. This seminar will explore the application of X-ray Absorption Spectroscopy (XAS) to identify spectral signatures and molecular structures of active sites under in situ and operando conditions. XAS is particularly advantageous for such studies due to its chemical selectivity and sensitivity to the local atomic environment, offering fundamental insights into the electronic and structural properties of the investigated materials.

A key challenge in catalysis for spectroscopists is solving the inverse problem consisting in recovering the structure of the active species from their XAS spectra, especially when multiple species contribute to the measured dataset. To address this, I will demonstrate how a combination of linear algebra and unsupervised machine learning (ML) techniques, such as Principal Component Analysis (PCA), can decompose XAS spectra into meaningful components. This approach allows for the extraction of pure spectra without the need for reference data, providing detailed insights into species speciation under varying conditions (e.g., temperature, gas feed, pH, etc.).

Furthermore, I will present how the catalyst atomistic structures can be refined from experimental data by combining supervised ML methods with ab initio calculations. A novel method will be introduced that integrates ML algorithms to extract structural information from both XANES (X-ray Absorption Near Edge Structure) and EXAFS (Extended X-ray Absorption Fine Structure) spectra, providing deeper insights into the molecular geometries of the active sites.

Finally, I will showcase successful applications of these advanced techniques. Specifically, I will discuss their role in understanding the structure of Cu-exchanged zeolites, which are critical for methane-to-methanol conversion and deNOx catalysis. I will also demonstrate their use in tracking dynamic changes in working Ni and Co nitrogen-doped carbon catalysts, which are emerging as efficient, earth-abundant materials for electrocatalytic CO₂ reduction.