Catalyst Diffusion Transformer: Generative Inverse Design of Heterogeneous Catalysts

2026-07-27

AI Summary

Traditional catalyst design is hindered by the vast chemical space and strongly coupled variables, making it challenging to efficiently generate materials with target properties in an inverse manner. This work proposes CatDiT, a framework that integrates diffusion models with Transformer architecture to enable multi-condition controllable generation—from intermetallic alloys to oxide surfaces—within a compressed latent space. CatDiT uniquely unifies reliable control over discrete attributes with directional tuning of continuous properties. The approach supports synergistic guidance by multiple conditions, including adsorbate type, binding energy, and catalyst class. Applied to the nitrogen reduction reaction, it successfully generates 28 DFT-relaxed alloy candidates that satisfy the activity window and break the conventional *N–*H scaling relations observed in pure metals, achieving approximately 1.5-fold enrichment over the source distribution.

Abstract

The vast chemical design space and complex, interdependent design variables make catalyst discovery for targeted properties highly labor- and resource-intensive. Although generative models have emerged as a promising solution, existing approaches are generally limited to single-property conditioning or narrow chemical spaces. Here, we present Catalyst Diffusion Transformer (CatDiT), a unified framework for inverse catalyst design that generates valid and novel structures ranging from intermetallic alloys to oxide surfaces. By learning compressed latent representations, CatDiT enables efficient training and rapid sampling while supporting simultaneous conditioning on adsorbate type, binding energy, and catalyst class. The model provides reliable control of discrete properties and directional control of continuous properties, enriching candidate pools for reaction-specific catalyst discovery. As a representative application, multi-conditional generation for the nitrogen reduction reaction (NRR) yields 28 density functional theory (DFT)-relaxed alloy candidates that satisfy the target activity window and lie above the pure-metal *N-*H scaling line, corresponding to a ~1.5-fold enrichment over the source distribution. These results establish CatDiT as a practical and scalable approach for property-directed catalyst inverse design and targeted catalyst generation.

Problem

• heterogeneous catalysts
• inverse design
• chemical design space
• property-directed discovery
• multi-conditional generation

Innovation

• Catalyst Diffusion Transformer
• inverse design
• generative modeling
• multi-conditional generation
• heterogeneous catalysts