QuantumATK

LCAO Total Energy Methods

  • Numerical atomic orbital basis sets with compact support
  • Optimized basis sets for most elements with low, medium and high accuracy
  • Norm-conserving Troullier-Martins pseudopotentials
    • FHI/SG15/PseudoDojo potentials provided for almost all elements of the periodic table, including semi-core potentials for many elements
    • PseudoDojo and SG15 potentials are fully relativistic
  • Over 300 LDA/GGA/MGGA exchange-correlation functionals via libXC
  • Methods for accurate band gap calculations of semiconductors and insulators
    • MetaGGA
    • DFT+1/2 method
    • Empirical "pseudopotential projector shift" method (parameters provided for Si and Ge)
  • Van der Waals models (DFT-D2 and DFT-D3)
  • Non-collinear, restricted and unrestricted (spin-polarized) calculations
  • Spin-orbit coupling
  • Hubbard U term in both LDA and GGA (also spin-dependent)
    • "Dual", "on-site", and "shell-wise" models
  • Counterpoise correction for basis set superposition errors (BSSE)
  • Ghost atoms (vacuum basis sets) for higher accuracy in the description of surfaces and vacancies
  • Virtual crystal approximation (VCA)
  • Analytical Force and Stress

PlaneWave Total Energy Methods

  • Plane-wave basis sets with default mesh cutoff setting for all elements
  • Hybrid functionals using the ACE approximation
  • Norm-conserving Troullier-Martins pseudopotentials
    • FHI/SG15/PseudoDojo potentials provided for almost all elements of the periodic table, including semi-core potentials for many elements
    • PseudoDojo and SG15 potentials are fully relativistic
  • Over 300 LDA/GGA/MGGA exchange-correlation functionals via libXC
  • Methods for accurate band gap calculations of semiconductors and insulators
    • MetaGGA
    • DFT+1/2 method
    • Empirical "pseudopotential projector shift" method (parameters provided for Si and Ge)
  • van der Waals models (DFT-D2 and DFT-D3)
  • Non-collinear, restricted and unrestricted (spin-polarized) calculations
  • Spin-orbit coupling
  • Eigensolvers
    • Default: Generalized Davidson method – stable and robust method
    • Also includes (as the first code) the Projected Preconditioned Conjugate Gradient (PPCG) algorithm for computing many extreme eigenpairs of a Hermitian matrix (cf. https://arxiv.org/abs/1407.7506 & https://doi.org/10.1016/j.jcp.2015.02.030); provides better scaling for large scale parallel calculations (single k-point, many processes per k-point)
  • Analytical Force and Stress

Semi Empirical Total Energy Methods

  • DFTB-type model, 30 different parameter sets are shipped with the product, and more can be downloaded and used directly
  • Built-in Slater-Koster models for group IV and III-V semiconductors (including strained models)
    • Interface for input of user-defined Slater-Koster parameters
  • Extended Hückel model with over 300 basis sets for (almost) every element in the periodic table
  • Tight-Binding models for strained systems (T. B. Boykin et al., Phys. Rev. B 81, 125202 (2010))
  • Spin polarization term can be added via internal database of spin-split parameters
  • Non-collinear spin
  • Spin-orbit interaction (parameterized)
  • Hartree term for self-consistent response to the electrostatic environment
  • All models adapted for self-consistent calculations through external database of atomic Hartree terms following the DFTB approach
  • Analytical Force and Stress

ForceField Total Energy Methods

  • Over 300  empirical classical potentials included
    • Two/three-body potentials: Lennard-Jones (various versions), Coulomb (various versions), Stillinger-Weber, Tersoff (various versions), Brenner, Morse, Buckingham, Vessal, Tosi-Fumi, user-defined tabulated
    • Many-body: EAM, MEAM, Finnis-Sinclair, Sutton-Chen, charge-optimized many-body (COMB)
    • Polarizable: Madden/Tangney-Scandolo, core-shell
    • ReaxFF
    • Valence force field (VFF) models
  • BYOP (Bring Your Own Potential)
    • Python interface for adding your own or literature potential of any of the supported types
    • Support for custom combinations of potentials
      • E.g. use a Stillinger-Weber potential with a Lennard-Jones term to account for van der Waals interaction
      • Several such potentials from literature are already provided: Pedone, Guillot-Sator, Marian-Gastreich, Feuston-Garofalini, Matsui, Leinenweber, Madden, and more
  • Coulomb solvers
    •  Ewald (smooth particle mesh), DSF, Debye, simple pairwise
  • Interface for adding your own or literature potential of any of the above types
  • Parallelized via OpenMP for optimal multicore performance (MPI parallelization in implementation) 

Ion Dynamics for LCAO, PlaneWave, SemiEmpirical and ForceField

  • Quasi-Newton LBFGS and FIRE methods for geometry and unit cell optimization (forces and stress)
    • Simultaneous optimization of forces and stress
    • Optimize structure to specified target stress (hydrostatic or tensor)
    • Pre/post step hooks for custom on-the-fly analysis
  • Computation of dynamical matrix
    • Phonon band structure, DOS, and thermal transport
    • Compute and visualize phonon vibration modes
    • Compute the Seebeck coefficient, ZT, and other thermal transport properties by combining ionic and electronic results
    • Zero-point energy and free lattice energy can be obtained from the PhononDensityOfStates analysis object (vibrational free energy in quasi-harmonic approximation of molecules and bulk)
    • Wigner-Seitz approximation for large supercells
  • Geometry optimization of device structures (also under finite source–drain bias)
  • Calculation of transition states, reaction pathways, and energies
    • Nudged elastic bands (NEB) method, enhanced version developed in-house
    • Support for varying cell shape and size, to simulate e.g. phase changes
    • Climbing image method
    • Pre-optimized path using the image-dependent pair potential (IDPP) (http://scitation.aip.org/content/aip/journal/jcp/140/21/10.1063/1.4878664)
    • Parallelized over images
  • Molecular Dynamics (MD)
    • State-of-the-art MD dynamics methods
      • Runs with DFT, semi-empirical models, or classical potentials
      • All thermostats and barostats support linear heating and cooling
      • All barostats support isotropic and anisotropic pressure coupling and linear compression
    • Thermostats and barostats
      • NPT with stress mask
      • NVT Nosé-Hoover with chains
      • NVE Velocity Verlet
      • NVT/NPT Berendsen
      • Martyna-Tobias-Klein barostat
      • Langevin
    • Several options for initialization of velocities
    • Pre/post step hooks in Python for custom on-the-fly analysis or custom constraints
  • Flexible constraints
    • Fix atoms
    • Separate X, Y, Z constraints
    • Fix center of mass in MD
    • Constrain Bravias lattice type (even when target stress is applied)
  • Partial charge analysis
  • Visualization of velocities
  • Interactive analysis tool for trajectory and single configuration properties (also for imported trajectories from LAMMPS, VASP, etc)
    • radial/angular distribution function
    • velocity autocorrelation
    • local mass density profile
    • coordination number
    • mean-square displacement
    • nearest neighbor number
    • neutron scattering factor
    • velocity/kinetic energy distribution
    • local structure analysis (Voronoi type)
    • centrosymmetry
    • In scripting, the above analysis can be performed very efficiently for a selected subset of atoms, also in very large structures
  • Mechanical properties
    • Forces and stress (analytic Hellmann–Feynman)
    • Elastic constants
    • Local stress
  • Global optimization 
    • Genetic algorithm for crystal structure prediction
  • Adaptive Kinetic Monte Carlo (AKMC)
    • Long time scale molecular dynamics for finding unknown reaction mechanisms and estimating reaction rates
  • Harmonic transition state theory (HTST) analysis of transition rates
    • Two options: detailed analysis via phonon partition function, or quick estimate via curvature of NEB path
    • Metadynamics via the PLUMED library (http://www.plumed.org/)
  • Export movies of MD trajectories, phonon vibrations, NEB paths, etc.

Poisson Equation Solvers for LCAO, PlaneWave and Semi Empirical

  • FFT (for periodic systems)
  • Two solvers for systems including metallic/dielectric regions:
    • Multigrid
    • Conjugate gradient method (parallelized in memory and execution)
  • FFT2D solver for transport systems with different dipoles (no need for dipole correlations)
  • "Direct" solver for large-scale calculations (parallelized in memory)
  • Multipole expansion for molecules
  • Dirchlet, von Neumann, or periodic boundary conditions can be specified independently in each direction
  • Metallic gate electrodes and dielectric screening regions
    • Allows for computation of transistor characteristics (gated structures) as well as charge stability diagrams of single-electron transistors

Performance Options for LCAO, PlaneWave and Semi Empirical

  • Consistent use of "best in class" standard libraries/algorithms like Intel MKL, ELPA, PETSc, SLEPc, ZMUMPS and FEAST
  • Proprietary sparse matrix library
  • Parallel memory distribution of e.g. the mixing history
  • Automatic adjustment of number of bands above the Fermi level to include
  • Multilevel parallelism
    • Over images in NEB and similarly for other complex tasks
    • Over k-points
    • Over basis functions (using multiple processes per k-point)
    • Also for band structure, DOS etc.
    • Automatic algorithm to determine the default (optimal) number of processes per k-point
  • Caching of data for higher memory usage vs. faster performance - or opposite
  • Use disk space instead of RAM to store grids for Poisson solver
  • PEXSI solver for O(N) calculations of very large systems (10,000+ atoms in DFT); cf. http://arxiv.org/abs/1405.0194
  •  Automatic threading intelligence
    • the number of threads is automatically controlled to make the best use of the available resources, taking into account the number of MPI processes used
  • Possible to combine MPI and openMP parallelization

Electronic Structure Analysis for LCAO, PlaneWave and Semi Empirical

  • Band structure
    • User defined Brillouin zone path through selection of high symmetry points
    • Fat bandstructure, shows projection onto atoms, spin, orbitals or angular momenta, in any desired combination
    • Effective bandstructure, i.e. unfolding of bandstructure for alloys and other supercells
  • Molecular spectrum
    • One-electron spectrum of molecules
    • Projected Gamma-point molecular spectrum for periodic systems
  • Density of states (DOS) 
    • Calculated using the tetrahedron method of gaussian smearing
    • Projection onto atoms, spin, orbitals or angular momenta, in any desired combination
    • Projections of band structure and DOS onto atoms, spin, orbitals or angular momenta, in any desired combination
  • Mulliken populations of atoms, bonds and orbitals 
  • Real-space 3D grid quantities as Python objects allowing for manipulations, evaluation at points,
    • Electron density
    • Effective potential
    • Full Hartree or Hartree difference potential
    • Exchange-correlation potential
    • Full electrostatic or electrostatic difference potential
    • Electron localization function (ELF) 
    • Molecular orbitals 
    • Bloch functions, complex wavefunction with phase information 
  • Total/free energy
    • Entropy contribution
  • Polarization and piezoelectric tensor
    • Calculated using the Berry phase approach
    • Calculation of Born effective changes
    • Optional internal ion relaxation
  • Effective mass analysis calculated using Finite Difference or Perturbation Theory 
    • 2nd order perturbation theory or analytic tensor
  • Bader charges 
  • Born effective charges
  • Fermi surface
  • Effective band structure (zone unfolding for supercells)
  • Optical properties 
    • Kubo-Greenwood formalism for linear optical properties
    • Calculation of optical adsorption, dielectric function, refractive index, etc.
  • Local Bandstructure
  • Effective mass
  • Charge point defect study object
    • A framework for studying point defects in bulk materials: vacancies, substitutionals, interstitials
    • Neutral and charged defects
    • Calculate relaxed defect structures, formation energies and thermodynamic transition levels
    • Can use with the new PlaneWave calculator
    • Can specify different calculators for the relaxation and the calculation of formation energies
    • Ghost atoms enabled for vacancy sites, also during relaxation, for significantly better accuracy
    • FNV correction scheme for charged defects with automatic Gaussian model charge fitting
    • Check convergence of defect properties with cell size and extrapolate to infinite size
    • Automatic adjustment of k-point sampling for different cell sizes to ensure best possible consistency between sizes

Additional Electronic Structure Analysis for LCAO and Semi Empirical

  • Complex band structure
  • Bulk transmission spectrum
  • Heisenberg exchange analysis module
    • A novel method of computing exchange coupling constants for the Heisenberg model, using the Green's-function method
    • Works within the framework of density functional theory (DFT+U) combined with the LCAO basis set approach
    • The Heisenberg spin-lattice model is an empirical approach to study various magnetic properties at finite temperatures, e.g., to understand phase diagrams, phase transitions, and magnetization dynamics of the magnetic system
    • The main difference from the traditional approach, which is based on total energy calculations for multiple magnetic configurations, is that all the exchange coupling constants can be computed from a single magnetic configuration calculation
    • Easy to set up the HeisenbergExchange analysis such that only couplings between certain atoms are considered
  • Spin life time
    •  At technologically relevant temperatures (>100 K) the spin life time will be limited by electron-phonon interactions, mediated by spin-orbit coupling (Elliot-Yafet mechanism)
    • QuantumATK 2018.06 can calculate the phonon-limited spin life time from an ElectronPhononCoupling object (if computed with noncollinear spin and spin-orbit coupling)

Special Features for LCAO, PlaneWave and Semi Empirical

  • Initialization of a new calculation via the self-consistent density matrix of a converged one (with automatic spin realignment)
  • Initialization of noncollinear spin calculations from collinear or spin-unpolarized ones for improved convergence
  • Custom initial spin-filling schemes
  • Odd/even k-point grids (Monkhorst-Pack or edge-to-edge zone filling), Gamma-centered or with custom shifts
  • Fractional hydrogen pseudopotentials and basis sets (for surface passivation)
  • Low-level interface to extract Green's function, Hamiltonian, overlap matrices, self-energies, etc.
  • Delta test module for benchmark of pseudopotential/basis set accuracy
  • Flexible and customizable verbosity framework to control the level of output to the log files
  • Region-dependent "c" parameter for TB09 Mega-GGA
  • Occupation functions: Fermi, Methfessel-Paxton, Gaussian, ColdSmearing
  • Local atomic shifts
  • Simulate external fields
  • Implicit solvent model
  • Support for charged systems
  • Compensation charges
    • Mimic charge doping
    • Passivate surface atoms 

NEGF for LCAO and Semi Empirical

  • NEGF method for two-probe systems
    • Non-equilibrium Green's function (NEGF) description of the electron distribution in the scattering region, with self-energy coupling to two semi-infinite leads (source/drain electrodes)
    • Open boundary conditions (Dirichlet/Dirichlet) allows application of finite bias between source and drain for calculation of I-V curve
    • Includes all spill-in contributions for density and matrix elements
    • Use of electronic free energy instead of total energy, as appropriate for open systems
    • Ability to treat two-probe systems with different electrodes (enables studies of single interfaces like metal-semiconductor or p-n junctions, for instance)
    • Ability to add electrostatic gates for transistor characteristics
  • Surface Green's function method for single surfaces
    • NEGF description of the surface layers, with self-energy coupling to a semi-infinite substrate (replaces the slab approximation with a more physically correct description of surfaces)
    • Appropriate boundary conditions for infinite substrate and infinite vacuum above the surface, both for zero and finite applied bias on the surface
  • Performance and stability options
    • Scattering states method for fast contour integration in non-equilibrium (finite bias)
    • O(N) Green’s function calculation and sparse matrix description of central region
    • Double or single semi-circle contour integration for maximum stability at finite bias
    • Ozaki contour integration to capture deep states
    • Sparse self-energy methods to save memory
    • Options to store self-energies to disk, either during calculation (instead of RAM) or permanently, to reuse in other calculations
    • Adaptive (non-regular) k-point integration for transmission coefficients
    • Minimal Electrode Concept
      • Reduced electrode - automatically repeated for computing self-energies
      • Saves time in the electrode calculation which is O(N3)
  • Calculation of I-V curves
    • Elastic, coherent tunneling transport
    • IV Characteristics Study Object
      • Combined framework for running multiple source-drain/gate voltage calculations and collecting and analyzing the results
      • Multilevel parallelism
      • “Smart restart“
      • Custom settings for individual voltages
      • State persistence (via data in HDF5)
      • Custom drain-source voltages, plot current as function of gate source; current as function of drain-source for one or many gate voltages
      • Show on/off ratio, subthreshold slope, transconductance, DIBL, source-drain saturation voltage
    • Quasi-inelastic (LOE) and fully inelastic (XLOE) electron-phonon scattering
      • Works with any combination of methods for the electronic and ionic degrees of freedom (DFT, tight-binding, DFTB, classical potentials)
      • Many performance options, such as averaging over phonon modes (bunching), using energy-dependent relaxation energies, and repeating the density matrix for homogeneous systems
      • Inelastic transmission spectrum (IETS) analysis
    • Special thermal displacement (STD) approximation to efficiently capture the effect of phonon scattering on the I-V curve by creating a canonical average over all phonon modes. For more details, see arXiv:1706.09290
  • Photocurrent Module
    • New analysis module for calculating the photocurrent and photon-mediated transmission in a device using first-order perturbation theory within the 1st Born approximation
    • Also gives the total current based on illumination by the AM1.5 standard solar spectrum
  • Device Configuration Study Object for Relaxation of Devices
    • This study object fully automates the Bulk Rigid Relaxation (BRR) method, making it very simple to optimize the geometry of a device
    • Can relax also complex device systems
  • Analysis of transport mechanisms
    • Transmission coefficients (k-point/energy resolved)
    • Monkhorst-Pack or edge-to-edge zone filling k-point scheme, or sample only part of the Brillouin zone for detailed information
    • Spectral current
    • Transmission spectrum, eigenvalues, and eigenchannels
    • Device density of states, also projected on atoms and angular momenta
    • Voltage drop
    • Molecular projected self-consistent Hamiltonian (MPSH) eigenvalues
    • Current density and transmission pathways
    • Spin-torque transfer (STT) for collinear/non-collinear spin
    • Atomic-scale band diagram analysis via LDOS or device DOS

Electron-Phonon Interaction for LCAO and Semi Empirical

  • Extract electron-phonon coupling matrix elements
  • Compute deformation potentials and conductivity/mobility tensor, via the Boltzmann equation, with constant, k-point and/or only energy-dependent relaxation times
  • Compute Hall coefficient and Hall conductivity tensor, Seebeck coefficient and ZT, first moment, and thermal conductance automated workflows for dynamical matrix (D) and Hamiltonian derivatives (dH/dR)
  • Wigner-Seitz approximation for calculations of both D and dH/dR
  • Tetrahedron integration method for calculating mobility and resistivity of states

NanoLab

  • Atomic geometry builder for molecules, crystals, nanostructures and devices
    • Symmetry information tool
    • Supercells
    • Interactive control of structure, select, edit, move, by atom, fragment, etc.
    • Surface cleaver 
      • Select Miller indices, surface Bravais Lattices and cleavage planes
      • Create slabs or supercell geometries 
    • Interface builder 
      • Analyze strain for different supercell sizes and crystal rotations
      • Optimize interface geometry
    • Icosahedron builder plugin
      • Build icosahedron nanoparticles
    • Wulff construction tool
      • Build nanoparticles with minimal surface energy
    • NEB tools
      • Set up path
      • Edit images collectively or individually
      • Pre-optimize NEB path with Image Dependent Pair Potentials (IDPP) 
      • Access interpolation algorithms  (LI-LinearInterpolation, HLC-HalgrenLipscomb, and IDDP-ImageDependentPairPotential) in Python scripts for easier automation of NEB path generation
    • Create device structures for transport calculations
    • Builders for nanostructures like graphene, nanotubes, nanowires
    • Molecular builder
    • Polycrystalline builder
    • Passivation tool for surfaces to remove bonds 
    • Import/export of most common atomic-scale modeling file formats (extendable by plugins; embedded version of OpenBabel)
    • Packmol plugin
    • Special Quasi-random Structures (SQS) algorithm for generating random alloys
      • Use a genetic algorithm (unlike other codes that perform an open-ended Monte Carlo simulation, which can be very slow)
      • Supports 2-component systems like SiGe or InGaAs
      • Any type of geometry, also nanowires etc.
    • Python Console
      • Provides direct Python access to interact with the configurations in the Builder
      • Maps (some) operations in the Builder to Python commands
      • Create pre-defined scripts (”snippets”) to automate repeated tasks
  • Databases
    • Internal structure library with several hundred basic molecules and crystal structures
    • Interface to query online databases such as
      • Crystallography Online Database (http://www.crystallography.net/cod/)
      • Materials Project
    • Support for custom, internal databases based on MongoDB or MySQL
  • Easy setup of calculations, even advanced work-flows
    • Full range of functionality for LCAO, PlaneWave, SemiEmpirical and FordeField
  • Viewer for 3D data
    • High-performance shader-based rendering engine for very large data sets (1M+ atoms and bonds)
    • Isosurfaces, isolines, and contour plots, with graphical repetition with data range control
    • Control atom color, size, transparency, etc.
    • Color atoms by computed quantities, like forces, velocities
    • Also works in movies, e.g. MD trajectories
    • Polyhedral rendering of crystals
    • Voxel plot (point cloud) rendering of 3D grids
    • Vector field plots
    • 3D extrusion of contour plans
    • 3D scene control, multiple light sources
    • Brillouin zone explorer
    • Export images in most common graphical formats
    • Export (and import) CUBE or simple xyz data files for external plotting
    • Export movies of MD trajectories, phonon vibrations, NEB paths, etc
    • Auto-rotated views can be exported as animated GIFs
    • Interactive 3D measurement tool for distances and angles
  • 2D plot framework
    • Save and reuse customized plots by converting plots to Python scripts
    • Combine plots, e.g. band structure and DOS
    • Add annotations like arrows and labels to plots
  • Project management
    • Organize data files into projects
    • Easily transfer projects between computers, or share with other users
    • Overview all data in a project, or focus on particular subsets, then combine data sets from different files for advanced analysis
  • Editor
    • Search-and-replace
    • Syntax highlighting
    • Python code completion
    • Select font
  • Job Manager
    • Submit and run jobs from the GUI in serial, using threading and in parallel using MPI
    • Submit jobs from the GUI to local machines
    • Submit jobs from the GUI to remote machines
      • A variety of queue types: Torque/PBS, LSF, SLURM, and direct execution (no queue)
      • Additional queue types can be added by plugins
      • Special plugin for QuantumATK On-Demand on Sabalcore
      • Requires only SSH access from client to server (no server-side daemon is required, all is controlled by the client)
      • Automatically copies input and output files to/from remote resources
      • Built-in SHH key generation and transfer to remote host (no need of 3rd party programs)
      • Diagnostics tool checks that added machine settings are correct
  • Python scripting interface, directly coupled to GUI
    • Can also be used interactively
    • Parallel scheduler
    • Includes PyQt4
    • PyMatGen included (pre-compiled)

Python Scripting and Automatization

Python Scripting is the component that binds all the calculators together in a common interface and allows them to synergistically work together. It also enables users to automate and customize task (also in NanoLab). 

atkpython is a python 2.7 interpreter with a large number of python modules preconfigured. It can run in interactive mode and in batch mode. The input format for QuantumATK is a python script which setup the simulation using native python commands together with QuantumATK python functions for

  •  Structure generation
    • Define molecule, bulk, surface and device geometries
    • Define Bravais lattices
    • Build special geometries like nanowires, graphene sheets, nanotubes
    • Reproduce workflows from the Nanolab builder using builder python commands
  • Simulation Setup 
    • Define simulation setup for QuantumATK DFT-LCAO, DFT-PlaneWave, TightBinding or ForceField
    • Define workflows which combine simulation engines
    • Add post or pre-hooks to Molecular Dynamics simulations, thereby tailoring the MD simulation algorithm
  • Post Analysis 
    • Automate analysis and plotting
    • Access internal QuantumATK variables for specialized analy
    • Batch processing of Analysis
    • Combine analysis of different simulations
  • There are more than 400 QuantumATK classes and functions available to the user, see list here
  • Variables are defined with physical units and QuantumATK allows for conversion between different units
    Units: nm, Ang, Bohr, Meter, Rydberg, eV, meV, Hartree, Joule, Calorie, kiloCaloriePerMol, kiloJoulePerMol, Newton, nanoNewton, kilogram, Kelvin, fs, femtoSecond, picoSecond, nanoSecond, microSecond, millisecond, Second, Minute, Hour, Day, Ampere, Volt, Siemens, G0, Coulomb, bar, Pa, Gpa, hbar, Mol, Radians, Degrees
  • Physical Constants    
     boltzmann_constant, planck_constant, Avogadro_number, speed_of_light, atomic_mass_unit, hbar, electron_mass, elementary_charge, vacuum_permitivity
  • 3rd party Python modules available from atkpython    
     ADODBAPI, ASE, CCLIB, Certifi, Colorama, Crypto, Decorator, H5py, Ipykernel, Ipython, Isapi, Jinja, Jupyter_core, markupsafe, matplotlib, monotonic,  monti, mpi4py, networkx, numpy, packaging, paramiko, pexpect, pickleshare, pillow, pkgconfig, plumed, psutil, pupynere, pybtex, pycrypto, pygments, pymatgen, pymongo, pymysql, PyOpenGL, PyQt5, pythonwin, pytz, pywin32, pyyaml, pyzmq, qtconsole, requests, scandir, scipy, simplegeneric, singledispatch, sip, six, spglib,  tornado, tabulate, traitlets, wcwidth, win_unicode_console, win32, win32com, zmq

Platform Support

  • Self-contained binary installer - no compilation needed, no external library dependencies beyond standard operating system packages
    • Support for all modern 64-bit Windows and Linux versions (detailed system requirements)
    • Provides a complete Python environment with precompiled optimized libraries like numpy/scipy/ScaLAPACK (based on MKL), matplotlib/pylab, Py4MPI, SSL bindings, Qt/PyQt, etc.
  • Parallelization (Windows/Linux)
    • QuantumATK is compiled against Intel MPI and the Intel Math Kernel Library (MKL) which in combination automatically provide an optimized balance between OpenMP threading and MPI
    • Intel MPI is included in the shipment
    • Support for MPICH2/MPICH3 (Ethernet), MVAPICH2 (Infiniband), and other MPICH-compatible libraries
  • Floating license system (SCL from Synopsys)