Accelerating Hierarchical ZSM‑5 Engineering via Bayesian Optimization-Guided Discovery.

Area of Science:

• Materials Science
• Chemical Engineering
• Catalysis

Background:

• Hierarchical zeolites, like ZSM-5, offer enhanced catalytic performance due to their dual pore systems.
• Optimizing the synthesis of hierarchical zeolites with controlled micro- and mesoporosity is challenging.
• Efficient synthesis methods are needed to minimize experimental effort and accelerate materials discovery.

Purpose of the Study:

• To accelerate the synthesis of hierarchical ZSM-5 with a balanced micro-mesoporous structure using Bayesian optimization.
• To identify optimal synthesis conditions for maximizing the hierarchy factor and improving catalytic properties.
• To establish a data-driven workflow for efficient zeolite design and optimization.

Main Methods:

• Bayesian optimization (BO) guided by Gaussian process regression was employed.
• 15 initial experiments informed three successive BO iterations.
• Characterization techniques included N2 physisorption, 27Al and 29Si NMR, and acidity analysis.

Main Results:

• The optimized sample (HZ-R0.55T50t2) achieved the highest hierarchy factor (0.17).
• The optimized zeolite exhibited high mesoporosity (0.44) and microporosity (0.38), indicating reduced diffusion limitations.
• NaOH and TPAOH synergistically created uniform mesopores while preserving the zeolite framework and crystallinity.
• Sensitivity analysis revealed TPAOH fraction and temperature as key factors influencing the hierarchy factor.

Conclusions:

• Bayesian optimization is an effective tool for accelerating the design and synthesis of hierarchical zeolites.
• The developed data-driven workflow minimizes experimental effort in optimizing complex material structures.
• The findings provide a pathway for designing advanced zeolite catalysts with tailored properties.