Finding Furfural Hydrogenation Catalysts via Predictive Modelling.

Area of Science:

• Catalysis
• Organic Chemistry
• Computational Chemistry

Background:

• Transfer hydrogenation is a key reaction in organic synthesis.
• Ruthenium-carbene complexes are promising catalysts but require optimization.
• Predictive modeling can accelerate catalyst discovery.

Purpose of the Study:

• To develop and validate a predictive model for ruthenium-carbene catalysts in transfer hydrogenation.
• To optimize catalyst performance for the conversion of furfural to furfurol.
• To demonstrate the utility of predictive modeling in catalyst design.

Main Methods:

• Synthesis and screening of 18 ruthenium-carbene complexes.
• Catalytic performance studies including yield, control experiments, and deuterium-labeling.
• Mechanistic investigations to determine the reaction pathway (monohydride pathway).
• Development of a predictive model using 2D and 3D molecular descriptors.
• Validation of the model through cross-validation and prediction for new catalysts.

Main Results:

• Ruthenium-carbene complexes showed varied yields (62% to >99.9%) in furfural transfer hydrogenation.
• Mechanistic studies confirmed the monohydride pathway and carbene ligand stability.
• A predictive model achieved high accuracy (R(2)=0.913) for catalyst performance.
• Predictions for new catalysts were within 3% of experimental results.

Conclusions:

• Predictive modeling is a valuable tool for optimizing transfer hydrogenation catalysts.
• The developed model accurately predicts the performance of novel ruthenium-carbene complexes.
• This approach accelerates the discovery and design of efficient catalytic systems.