Author Topic: QuantumATK U-2022.12 version released on Dec 6, 2022  (Read 6636 times)

0 Members and 1 Guest are viewing this topic.

Offline jessicaj

  • QuantumATK Staff
  • New QuantumATK user
  • *****
  • Posts: 2
  • Country: gb
  • Reputation: 0
    • View Profile
QuantumATK U-2022.12 version released on Dec 6, 2022
« on: December 6, 2022, 11:37 »
We are very pleased to announce QuantumATK U-2022.12! The latest version of the QuantumATK atomic-scale modeling platform includes many new and improved features and performance improvements.

If you are a customer entitled to maintenance services, you can download QuantumATK U-2022.12 installers, new license keys and full release notes directly from SolvNetPlus.

Highlights of the U-2022.12 Release:

Machine-Learned Force Fields for Realistic Structures and Thermal Properties
Defect-Assisted Recombination from DFT Calculations
Modelling of Homogeneous Electric Fields in DFT
Atomistic Spin Dynamics - Integration with Vampire
COSMO and COSMO-RS
R2SCAN Exchange-Correlation Functional
PAW method for LCAO
New Workflow Builder for Setting Up Simple and Complex Calculation Workflows
Updates Related to Molecular Dynamics and Force Fields
Various GUI & Platform Updates

Machine-Learned (ML) Force Fields for Realistic Structures and Thermal Properties

In this release, we have continued to develop our ML force field framework using Moment Tensor Potentials (MTPs). 1000-10,000x shorter computational time vs DFT enables ab initio accuracy for large system sizes and time scales greatly exceeding those accessible to DFT.

Available MTP library with pre-trained MTPs for a range of bulk materials and interfaces.

Updates to the Core MTP Framework

Significantly improved accuracy for multi-element systems using more a flexible design of the potential coefficients
New extrapolation grade algorithm for active learning: Query-by-Committee
The most important MTP training workflows are now accessible in the new Workflow Builder in the Nanolab GUI for fast and easy setup.
Built-in MTP analyzer to plot correlation between MTP and reference DFT energy, forces, and stress from Data Tool in Nanolab
Training protocol updates (generating training data for alloy materials and improvements for interface training protocol)
Improvements to MTP training functionality
CrystalPropertyValidation analysis object

Pretrained MTP library

New pre-trained MTP models
Re-trained MTP models for HKMG stack and interfaces, with added gate metals Ru, Sc, and their interface with HfO2
Re-trained TiNAlO potential with improved accuracy

Multilayer Stack Generation using the Multilayer Builder GUI

New MTJ builder to build stack configurations for magnetic RAM (MRAM) applications
Improvements in features and quality of the generated structures from the HKMG builder
General multilayer builder improvements, specifying displacement vectors for better aligning of crystal interfaces
Tutorials to explain how to build HKMG and MRAM stacks

Defect-Assisted Recombination from DFT Calculations

Support calculations of Shockley-Read-Hall (SRH) capture rates. SRH calculations are supported for LCAO and PlaneWave calculators, for both LDA/GGA functionals and hybrid HSE
Compute luminescence line shapes for color centers, which can be used for defect characterization by modelling photoemission experiments
Dedicated SRH Analyzer makes it easy to perform advanced analysis and drill down deeper into the results, and e.g., inspect the individual electron-phonon coupling contributions

Modelling of Homogeneous Electric Fields in DFT

Updates to electric fields implementation in periodic calculations (integrated and extended feature in terms of applicability)
The model accounts for the effects of field-driven electric enthalpy, forces, and stresses via the Berry-phase polarization theory
Supported for LCAO and Plane-Wave
Suitable for insulators and semiconductors (metals are not supported)
Provides a TAT/QoR trade off control: recalculation of Polarization and Born Effective
Charge along trajectories can be tuned from “frozen” to continuously updated
Performance improvements for Polarization and Born Effective Charge calculations and additional performance can be gained by using the newly enabled multilevel parallelism

Atomistic Spin Dynamics - Integration with Vampire

Vampire (atomistic magnetic simulation code) is integrated into workflows that enable studying spin dynamics and stability of the magnetic free layer in magnetic tunnel junctions for MRAM technology

This workflow is an ideal combination of simulation methods:
DFT-based input to spin dynamics calculations
Use ML-fitted force fields to obtain more than 1,000x faster geometries with the same accuracy
Vampire can compute a wide variety of properties with atomic resolution for realistic system sizes (5,000+ atoms)

Set up the full DFT-to-Vampire simulation flow in the NanoLab GUI, which automatically builds the required script to compute the required input parameters Heisenberg exchange constants and magnetic anisotropy energy
Launch the Vampire calculation from a QuantumATK Python script via wrapper classes (works both on Windows and Linux)
Visualize Vampire output Curie temperature, anisotropy energy, and hysteresis curves can be visualized and analyzed using dedicated plotting tools in the GUI
One can study STT-MRAM & SOT-MRAM, switching time & mode, temperature dependence,  skyrmions, 2D magnetic materials and much more. More information can be found in Vampire website

COSMO and COSMO-RS

A major new feature in this release is the inclusion of COSMO (COnductor-like Screening MOdel) to study solvation effects, as well as COSMO-RS to evaluate solubility and a range of other properties.

The calculations can be performed with the solvent either as a perfect conductor or as a dielectric corresponding to a particular solvent
Setting up COSMO calculations is automated in the new Workflow Builder and is added as an option in LCAO-DFT calculations for molecule and bulk configurations, and Device LCAO-DFT for surface configurations
Non-electrostatic effects such as dispersion and cavitation energies are estimated with a linear regression model
Solvation free energies, including non-electrostatic and gas phase terms can be automatically calculated
The COSMO solvent surface can be visualized in the Viewer, showing the solvent cavity shape and the charge polarization across the surface

COSMO-RS is an extension of the COSMO model that uses the surface charges generated in the COSMO calculation to calculate the chemical potential of a molecule. GUI-features of the COSMO-RS Analyzer include:

Calculations of different properties can be set up easily, including investigating the effects of temperature or different solvent composition
Results are shown graphically and in text format
The sigma profile and potential for molecules can also be displayed
The analyzer includes a database of approximately 1,500 molecules that can be added to a COSMO-RS calculation
New molecules can be added to the database through a simple drag and drop interface. The appropriate DFT calculations for each molecule can be automatically set up in the Workflow Builder
The parameters of the COSMO-RS method are editable in the COSMO-RS Analyzer. Some pre-defined parameter sets can be selected, or a user-defined parameter set can be used. Parameter sets can also be saved for later use

R2SCAN Exchange-Correlation Functional

The recently developed r2SCAN meta-GGA functional is now available in QuantumATK and can be simply selected in the GUI Scripter.
r2SCAN maintains the numerical performance that rSCAN provides over SCAN, but unlike rSCAN it satisfies the exact constraints that the exchange-correlation should obey.

PAW method for LCAO

LCAO-DFT calculations can now also make use of the PAW (Projector Augmented Wave) method, as has been possible with PlaneWave-DFT for some time. Note that in the U-2022.12 release, not all analysis options are yet available for LCAO-PAW, support is limited to bulk and molecule configurations, and only GGA functionals can be used
Optimized basis sets are available for Fast and Accurate calculations

New Workflow Builder for Setting Up Simple and Complex Calculation Workflows

The Workflow Builder is a tool to set up complex workflows which can involve several separate simulations which are later used as input for various post-processing calculations. Here is a list of noteworthy features:

When editing Workflow Blocks, non-default values are clearly indicated, and there is a function for direct "reset" to defaults
All modified Workflow Blocks can be saved in the "Blocks" catalog for easy reuse in other workflows
Workflow Blocks can be grouped together in a container block called the "Block of Blocks". Since a Block of Blocks can be saved, this allows for the re-use of even very complex calculation setups
The Workflow Builder supports the inclusion of already calculated results through the "Load Block". All QuantumATK objects can be loaded and used directly within a workflow. The obvious example is the calculation of additional analyses from an already updated configuration
When exporting to the script, the Workflow Builder allows for specifying both filename and object id

Updates Related to Molecular Dynamics and Force Fields
Surface Process Simulations

Adaptive thermostat height enables efficient substrate thermalization even for thick layer growth

Polymer Simulations

Apply bead definitions to an all-atom structure to automatically convert it to a coarse-grain representation

Ion Dynamics

New state-of-the-art NVT thermostat (Bussi-Donadio-Parrinello stochastic velocity rescaling) which combines the advantages of the Berendsen and Nose-Hoover thermostats.
Added option to save raw (i.e., unconstrained) forces and stress in optimization trajectories to be able to use it for MTP training
The confusing parameter "disable_stress" in OptimizeGeometry has been replaced by "optimize_cell"

Force Fields

Included state-of-the-art methods for charge equilibration, e.g., D4-EEQ, which is based on the Grimme-D4 scheme
New force fields for various glass compositions
Additional ReaxFF parameter sets
New Stillinger-Weber parameters for AlGaN
New core-shell potential for InSnO (ITO)
Minor performance improvements for CPU and GPU versions

Various GUI Updates

Improved Graphics Support for Virtual Machines
The Data Tool, Builder, Workflow Builder, Viewer, and Database Tool windows are now embedded within the main interface, instead of opening as separate windows. Only the Editor and the Custom Scripter still open as windows
Added ability to extract and use custom quantities on Trajectory objects for coloring atoms in the Viewer
Updated Interface Builder
Enabled more precise drops in different available areas of a plot grid layout

Get QuantumATK U-2022.12

If you are a customer entitled to maintenance services, you can access QuantumATK U-2022.12 installers and new license keys directly from SolvNetPlus.

QuantumATK U-2022.12 release comes with significant licensing updates and every user who wants to run the new QuantumATK U-2022.12 version, will need to refresh the license file. Contact us or your license administrator for any question.

License keys can be found here: go to https://solvnet.synopsys.com/SmartKeys and then click ''Key Retrieval''.
« Last Edit: December 7, 2023, 15:06 by Vaida Arcisauskaite »