Synopsys QuantumATK 2025.06

Synopsys QuantumATK 2025.06

Synopsys QuantumATK is a comprehensive platform for atomistic simulations combining density functional theory (DFT), semi-empirical methods, and non-equilibrium Green’s function (NEGF) techniques. It enables predictive modeling of materials properties, electronic structure, and quantum transport from first principles. The platform bridges quantum-scale calculations with device-level simulations, making it essential for designing next-generation semiconductors, catalysts, batteries, and nanomaterials.

Primary Users

QuantumATK serves researchers and engineers at the atomic scale:

• Materials Scientists discovering new 2D materials, alloys, and nanostructures

• Semiconductor Researchers studying interfaces, defects, and novel channel materials

• Nanotechnology Engineers designing transistors, memristors, and quantum devices

• Computational Chemists investigating catalysis, surface reactions, and molecular electronics

• Device Physicists simulating electronic and thermal transport at the nanoscale

• Academic & National Lab Researchers in physics, chemistry, and materials science

Key Features & Capabilities

Core Simulation Engines

• Density Functional Theory (DFT): Full-featured DFT with advanced exchange-correlation functionals (LDA, GGA, meta-GGA, hybrid)

• Non-Equilibrium Green’s Function (NEGF): Quantum transport for nanoscale devices including coherent and dissipative transport

• Semi-Empirical Methods: DFTB (Density Functional Tight-Binding), extended Hückel for larger systems

• Classical Force Fields: ReaxFF, Tersoff, Brenner for molecular dynamics of complex systems

Specialized Modules

• Magnetic Materials: Spin-polarized DFT with non-collinear magnetism and Dzyaloshinskii-Moriya interactions

• Optical Properties: Time-dependent DFT (TDDFT) for excitation spectra and dielectric functions

• Thermal Transport: Phonon calculations, lattice thermal conductivity, and electron-phonon coupling

• Electrochemistry: Implicit/explicit solvation models for catalysis and battery materials

Workflow & Analysis

• Python Scripting: Full Python API for automated workflows and custom analysis

• Graphical Interface: NanoLab GUI for setup, visualization, and analysis

• Database Integration: Materials Project and other external database connectivity

• High-Throughput Computing: Automated workflows for materials screening

What’s New in QuantumATK 2025.06

Advanced Simulation Capabilities

1. Quantum Machine Learning (QML) Force Fields: AI/ML potentials trained on-the-fly during DFT calculations, enabling million-atom simulations with DFT accuracy

2. Quantum Computing Integration: Interface to quantum computers (IBM, Google) for selected quantum chemistry calculations

3. Exascale-Ready Algorithms: Enhanced scalability for Frontier and Aurora supercomputers (10,000+ GPUs)

4. Topological Materials Toolkit: Automated identification and characterization of topological insulators and semimetals

Enhanced Physics Models

5. Strongly Correlated Systems: Dynamical mean-field theory (DMFT) integration for heavy fermion and Mott insulator systems

6. Superconductivity: Migdal-Eliashberg theory for conventional superconductors

7. Spin-Orbitronics: Advanced spin-orbit coupling models for spin-charge conversion devices

8. Radiation Effects: Displacement damage and ion implantation simulations

Performance & Usability

9. NVIDIA Grace Hopper Support: Optimized for GH200 superchips with unified CPU-GPU memory

10. Federated Learning: Collaborative model training across institutions while preserving data privacy

11. Automated Workflow Discovery: AI suggests optimal simulation protocols based on target properties

12. Real-Time Collaboration: Shared simulation sessions with interactive visualization

🖥️ System Requirements

Linux (Primary Platform)

Minimum Requirements:

• OS: RHEL 8.8, Rocky Linux 8.8, Ubuntu 22.04 LTS

• CPU: Intel Xeon Gold 6326 or AMD EPYC 7313 (32 cores minimum)

• RAM: 128 GB (256 GB recommended for NEGF transport)

• GPU: NVIDIA A100 (40/80 GB) or H100 for GPU acceleration

• Storage: 500 GB NVMe SSD + 2 TB for scratch files

• MPI: OpenMPI 4.1+ or Intel MPI

Recommended HPC Configuration:

• OS: RHEL 9.2, SLES 15 SP4 HPC

• CPU: 4× AMD EPYC 9754 (512 cores total) or Intel Xeon Platinum 8592+

• RAM: 1-4 TB per node (for large DFTB/MD simulations)

• GPU: 8× NVIDIA H100 (640 GB total) or 4× GH200

• Storage: All-flash parallel file system (Lustre/DAOS) with 100+ TB

• Interconnect: NVIDIA Quantum-2 InfiniBand (400 Gb/s)

Windows (GUI/Pre-Post Processing)

Supported Configuration:

• OS: Windows 10/11 Pro 64-bit

• CPU: Intel Core i9-14900K or AMD Ryzen 9 7950X

• RAM: 128 GB minimum (256 GB recommended)

• GPU: NVIDIA RTX 4090 (24 GB VRAM) for visualization

• Storage: 1 TB NVMe SSD

• Note: Full calculations should be submitted to Linux servers