AFLOW

Official Resources

• Homepage: http://aflow.org/
• Documentation: http://aflow.org/documentation/
• Source Repository: http://aflow.org/install-aflow/ (Source available)
• License: GPL v3

Overview

AFLOW (Automatic Flow) is a prominent high-throughput ab initio calculation framework and database. It automates the generation, simulation, and analysis of materials using DFT (primarily VASP). AFLOW manages a massive database of calculated properties (AFLOWlib) and provides tools for convex hull construction, prototype generation, and machine learning (AFLOW-ML).

Scientific domain: High-throughput DFT, materials database, materials informatics
Target user community: Materials scientists, alloy designers

Capabilities (CRITICAL)

• Database: AFLOWlib contains >3.5 million materials entries.
• Automation: Manages DFT calculations, error correction, and symmetry analysis.
• Prototypes: Extensive library of crystal prototypes for structure generation.
• Properties: Electronic structure, thermal properties (AEL/AGL for elastic/Debye), vibrations.
• Machine Learning: AFLOW-ML API for predicting properties (gap, bulk modulus, etc.) using trained models.
• Symmtry: AFLOW-SYM for symmetry analysis.

Sources: AFLOW website, Comp. Mater. Sci. 58, 218 (2012)

Inputs & Outputs

• Input formats: aflow.in file, POSCAR
• Output data types: Database entries, properties JSON, web pages

Interfaces & Ecosystem

• VASP: Primary DFT engine
• Quantum ESPRESSO: Support available
• Web API: REST API for querying the database programmatically

Workflow and Usage

1. Generate aflow.in file from a structure or prototype.
2. Run aflow --run to execute the workflow (manages VASP).
3. Data is automatically post-processed and added to local or remote repository.
4. Query online database via API (aflow --search).

Performance Characteristics

• Highly scalable (runs on top supercomputers)
• Massive database of pre-computed binary and ternary alloys

Application Areas

• Alloy discovery (entropy stabilized alloys)
• Thermoelectric materials
• Superconductors
• Machine learning

Community and Support

• Developed by Curtarolo Group (Duke University)
• Active development
• Regular schools/workshops

Verification & Sources

Primary sources:

1. Homepage: http://aflow.org/
2. Publication: S. Curtarolo et al., Comp. Mater. Sci. 58, 218 (2012)

Confidence: VERIFIED

Verification status: ✅ VERIFIED

• Website: ACTIVE
• Documentation: COMPREHENSIVE
• Source: OPEN (GPL)
• Development: ACTIVE
• Applications: High-throughput DFT, database, ML