X2DB: Unifying Experimental and Computational 2D Materials Data to Accelerate AI-Driven Discovery

September 6, 2026

DTU Research Database Denmark

Overview

A new open infrastructure, X2DB, has been established to integrate experimental and computational data for 2D materials on a massive scale, centralizing previously dispersed knowledge. This robust database facilitates comprehensive classification, significantly improving the creation of balanced benchmark datasets for machine learning models and streamlining property screening for diverse applications. Notably, X2DB has identified over 270 unique 2D materials already synthesized in few-layer forms that also possess corresponding computational data.

In Depth

Background

Two-dimensional (2D) materials, exemplified by graphene, hold immense promise for next-generation electronics, sensors, energy devices, and catalysts, thanks to their unique physical and chemical properties. However, the discovery of optimal materials with specific functionalities from a vast theoretical space of millions of possibilities has traditionally been a challenging, time-consuming, and costly endeavor. Furthermore, fragmented data and low accessibility have posed significant bottlenecks to research progress. The establishment of integrated data infrastructures like X2DB marks a crucial milestone in materials informatics, poised to foster international research collaboration and dramatically enhance the efficiency of AI-driven materials discovery. This push for open science infrastructure, particularly championed by European institutions including DTU, aims to establish global leadership in data-driven scientific advancement.

Key Findings

This research introduces X2DB, an innovative open infrastructure designed for the large-scale integration of experimental and computational data on 2D materials. By centralizing previously dispersed knowledge, X2DB aims to accelerate scientific understanding and practical applications. This powerful database provides a balanced benchmark set essential for building and rigorously evaluating machine learning (ML) models, while also significantly streamlining the design of material property screening studies tailored for specific applications such as advanced electronics and energy storage.

Technically, X2DB seamlessly integrates experimental data—including synthesis conditions, property measurements, and stability assessments from laboratories worldwide—with computational data, such as electronic structure, lattice vibrations, and band gaps, derived from first-principles calculations like Density Functional Theory (DFT) using advanced algorithms. This integration yields comprehensive insights into material composition, structure, and a diverse array of physicochemical properties. A remarkable achievement of X2DB is the identification of over 270 unique 2D materials already synthesized in few-layer forms, which also have corresponding computational data available in the existing Computational 2D Materials Database (C2DB). This critical bridge between theory and experiment establishes a robust foundation, enabling ML models to make more realistic and accurate predictions. X2DB is designed for easy accessibility by data scientists and materials researchers through a dedicated API, promoting widespread adoption and utility.

Looking ahead, large-scale data integration platforms like X2DB are set to become indispensable for shaping the future of 2D materials research. The database is expected to expand continuously, incorporating richer data on diverse material properties and synthesis conditions. This expansion will empower ML models to achieve even higher predictive accuracy and sophisticated inverse design capabilities, eventually enabling AI to autonomously propose 2D materials optimized for specific application needs. Ultimately, X2DB could integrate with autonomous laboratories, evolving into the core of a ‘closed-loop discovery system’ that seamlessly combines data generation, model learning, and experimental validation. This transformative initiative promises to dramatically accelerate the practical application of 2D materials, significantly contributing to the realization of high-performance flexible electronics, ultra-efficient solar cells, and groundbreaking quantum devices, fundamentally transforming the material discovery process and increasing the pace of scientific breakthroughs by orders of magnitude on a global scale.

Source: https://orbit.dtu.dk/files/436723233/large-scale-integration-of-experimental-and-computational-data-for-2d-materials.pdf

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.