CatRetriever: Contrastive Representation Learning for Slab-to-Bulk Retrieval in Generative Catalyst Discovery

2026-07-13

AI Summary

This work addresses the challenge in generative catalyst design that surface slab models often lack corresponding bulk structures, making it difficult to evaluate bulk-dependent properties. To bridge this gap, the study introduces contrastive representation learning for the first time to the slab-to-bulk association task, constructing a shared latent space between surface slabs and bulk crystals to enable efficient bulk structure retrieval. The proposed generation-retrieval joint framework integrates crystal structure embeddings, nearest-neighbor search, and generative search expansion, facilitating simultaneous filtering based on structural compatibility and target adsorption energy ranges. The method achieves high retrieval accuracy, with R@1 exceeding 91% on in-distribution test sets and R@3 surpassing 98% on held-out test sets, thereby effectively enabling adsorption-energy-guided discovery of bulk catalysts.

Abstract

Inverse design is an emerging data-driven paradigm for efficiently navigating vast chemical spaces to discover new materials with targeted properties, and in the context of heterogeneous catalysis, surface generative models have recently advanced this goal by directly generating catalyst surface-adsorbate structures. However, these models typically operate at the slab level and do not provide the corresponding parent bulk structure, making it difficult to assess bulk-dependent properties such as formation energy, surface energy, crystallographic symmetry, and synthesizability. Here, we address this missing slab-to-bulk connection as a retrieval problem and introduce CatRetriever, a contrastive representation learning model that aligns slab and bulk crystal representations in a shared latent space. From a slab query, CatRetriever accurately retrieves plausible parent bulk candidates with R@1 > 91% and R@3 > 98% on both the in-distribution and holdout evaluation sets. We further extend the CatRetriever framework into an adsorption energy targeted bulk discovery pipeline that combines bulk retrieval, generative search space expansion, and adsorption energy distribution analysis. This workflow evaluates candidates by both structural compatibility with the query slab and their ability to access the target adsorption energy range across diverse surface environments. CatRetriever therefore provides a scalable route for connecting catalyst generative models with physically plausible and adsorption energy compatible bulk catalyst discovery.

Problem

• slab-to-bulk retrieval
• generative catalyst discovery
• heterogeneous catalysis
• inverse design
• bulk structure

Innovation

• contrastive representation learning
• slab-to-bulk retrieval
• generative catalyst discovery
• adsorption energy targeting
• crystal structure alignment