top of page
  • Katherine Hollingsworth

MedeA Software Release MedeA 3.3 -- Material Intelligence!

Materials Design, Inc. Announces the Release of MedeA 3.3, 'Material Intelligence'. "We are delighted to announce the release of MedeA 3.3," said Clive M. Freeman, CEO of Materials Design. “The MedeA Environment provides world leading materials simulation capabilities and MedeA 3.3 opens exceptional opportunities for the generation of materials insights and understanding.” With the integrated and intuitive MedeA Environment, researchers build quantum mechanical descriptions of systems using VASP 6, create Machine Learning based potentials to capture the resulting energy surface, employ these descriptions in large length and time scale simulations using the LAMMPS package, and analyse results interactively. MedeA 3.3 includes updates to key server codes including: VASP 6.2.1, LAMMPS29Oct20, and GIBBS 9.7.4, increased platform support, and a range of performance enhancements and productivity features throughout. Additional enhancements facilitate the evaluation of vibrational entropy contributions to system free energy, model construction and editing improvements, and enhanced and extended parameter sets for atomistic organic (PCFF+) and mesoscopic systems. Dr. Freeman added: "The release of MedeA 3.3 reflects detailed input from customers and careful development and validation work from Materials Design. There are new features in this release which extend simulation for many different material types and phases of matter. The Machine Learning features of MedeA are unique and powerful.” With full access to all underlying information, MedeA expedites the creation of Machine Learning Potential models, which can be employed in simulating substantial system length and time scales, with high fidelity to quantum mechanical accuracy at greatly reduced computational cost, and significantly reduced learning cost from a simulator's point of view. MedeA 3.3 is a truly exceptional development in materials simulation. Materials modelers, scientists, and engineers can now access machine learning based simulations efficiently.


Description of MedeA 3.3 New Features and Enhancements

1. Builders and Editors:

  • Edit cell

    • Added option to switch between standard cell orientation and a preset arbitrary orientation

  • Supercell builder

    • Option to rotate to standard orientation (default) or to preserve orientation according to cell construction

  • Thermosets

    • Faster re-assignment of atom types (selective reassignment)

    • Increased the number of sites supported in multi-sites crosslinking

  • Subsets

    • Definition of subsets based on position criteria

    • Extensions on the ability to transfer subset information upon system editing (copy/paste, merge, create supercells etc)

    • Additional capability of freezing subsets via the Subset Manager stage

    • Calculation of bond orders along with bonds

    • Export to mol2 format for non-periodic and periodic systems

    • Enhancements to exporting to pdb format

2. Engines:

  • VASP

    • VASP 6.2.1 executables

    • Automated procedure to calculate accurate optical spectra including excitonic effects via the Time Evolution approach

    • Enhanced robustness of energy of formation and optical spectra simulations


    • Updated to the LAMMPS29Oct20 official stable version

    • Supports NVIDIA GPUs on Windows

    • Supports running on OpenMP threads on Linux and Windows

    • Supports the PLUMED package on Linux


    • GIBBS 9.7.4

    • Improved handling of simulation output (configurations in pdb formats)

    • Extended use of variables as simulation input (ideal gas heat capacity)

    • Improvements on printed simulations' summary in Job.out

3. Forcefields:

  • MLP (Machine Learning Potentials)

    • Access to a variety of published machine learning potentials for use with MedeA LAMMPS

  • MLPG (Machine Learing Potential Generator)

    • Generation of machine learning potentials (on Linux and Windows)

  • Martini

    • Addition of recently published Martini 3 for CG simulations

  • ForceField Optimizer (FFO)

    • Enhanced support for fitting force and stress data

    • Enhanced support for triclinic cells

    • Improved performance for some forcefield classes

4. Property Modules:

  • Phonon and MT

    • Enable automatic evaluation of the energy of formation for the system by MedeA VASP 6, thus providing temperature dependent energy of formation, entropy and free energy of formation

  • TSS

    • Enable automatic evaluation of the energy of formation by MedeA VASP 6 for initial and final configuration, as well as for all optimized transition states

    • Various enhancements for performance

  • Electronics

    • Enable automatic evaluation of the energy of formation for the investigated system by MedeA VASP 6

  • P3C

    • Improvements in the calculation of 4 descriptors (Namide, Nco, Nhb, Nvkh) used in the calculation of Ecoh1 and Ecoh2

New Modules:



bottom of page