Exploring Enzyme-Mimicking Metal-Organic Frameworks for CO₂ Conversion through Vibrational Spectra-Based Machine...

Area of Science:

• Materials Science
• Catalysis
• Computational Chemistry

Background:

• Natural enzymes offer efficient catalysis but have limitations.
• Metal-organic frameworks (MOFs) are highly tunable materials suitable for biomimetic applications.
• Carbonic anhydrase (CA) is a key enzyme for CO2 conversion.

Purpose of the Study:

• To screen a large library of Zn-MOFs for CO2 conversion capabilities.
• To identify MOFs that mimic the catalytic properties of carbonic anhydrase (CA).
• To explore structure-property relationships using vibrational spectral features.

Main Methods:

• Unsupervised machine learning was applied to density functional theory (DFT)-computed vibrational infrared and Raman spectral features.
• 300 different Zinc-based MOFs (Zn-MOFs) were screened.
• Spectroscopic attributes were compared to those of carbonic anhydrase (CA).

Main Results:

• MOFs exhibiting vibrational spectral features similar to CA demonstrated potential for replicating CA's electronic and catalytic properties.
• This approach utilizes vibrational spectral features, offering a more accessible method for exploring structure-property relationships compared to geometric configurations.
• Vibrational spectral features were identified as efficient carriers of chemical information, enabling multi-parameter optimization.

Conclusions:

• This study successfully screened 300 Zn-MOFs for CO2 conversion using machine learning and vibrational spectroscopy.
• The findings suggest that MOFs with CA-like vibrational spectra can mimic enzyme activity.
• This research advances the design of enzyme-mimetic MOFs and broadens the use of spectroscopy in biomimetic catalysis.