Virtual Kinetics Lab is a suite of open-source tools that enable multiscale modeling workflow to calculate thermodynamic properties of adsorbates on catalysts, reaction rate constants, reaction pathways and networks, kinetic models, and reaction model and kinetics visualization and analysis.

In this workshop we will present the theory behind the application of multiscale modeling for catalysis and reaction engineering. We will also demonstrate some of the key VLab software to simulate thermochemistry (pMuTT, pGrAdd) and microkinetic models (OpenMKM); visualize reaction network diagrams (ReNView); perform Bayesian design of experiments (NEXTorch); estimate model parameters (petBOA); investigate molecular similarity (AIMSim); and enable a data management framework for the entire workflow (CKineticsDB).

The workshop will be conducted as a webinar hosted on a cloud meeting platform. There will be live demonstrations of selected tools. The workshop is intended for scientists, researchers, and students interested in multiscale modeling. No prior experience is necessary to attend the workshop.

The source-code and documentation for the tools presented in this workshop is available on Vlachos Group GitHub and UD’s RAPID Software Website. The attendees will also be provided installation instructions for the tools and the code examples used for the demos closer to the workshop. However, we do not expect the attendees to work through the examples during the demo. They are encouraged to work through the examples during the week of the workshop after the presentations, so that any queries can be immediately shared with the organizers for feedback and further discussion.

Required Background Knowledge: None

Beneficial Background Knowledge: Basic knowledge of chemical kinetics, Python, basic knowledge of Docker to run OpenMKM containers.

Required Software: Any Operating System (Windows/Mac/Linux) which can run Python Jupyter notebooks and Docker for running OpenMKM. If Docker is unavailable, OpenMKM can be compiled from source but is not recommended for beginners.

**An overview of each tool can be accessed by topic below.

VLab provides several tools to accelerate multiscale modeling workflows involving design of experiments, parameter estimation, and uncertainty quantification (UQ). A short summary of the key tools is provided here.

Next Experiment Toolkit in PyTorch (NEXTorch) facilitates experimental design using Bayesian Optimization (BO), supports both automated and human-in-the-loop optimization, and offers various visualization options. It can be used to design laboratory experiments, molecular modeling simulations, reaction condition optimization, reactor geometry optimization, and CFD/multi-physics simulations, without extensive programming effort so that the user can focus on domain-specific questions. NEXTorch is built upon the backend BoTorch which enables GPU acceleration, parallelization, and state-of-the-art Bayesian optimization algorithms.

The Chemical Kinetic Bayesian Inference Toolbox (CKBIT) facilitates Bayesian inference upon kinetic model parameters. Bayesian techniques estimate optimal parameter values (maximum a posteriori) or probability distributions (Markov chain Monte Carlo, variational inference) rapidly. With minimal coding, CKBIT enables users to estimate activation energies, pre-exponential terms, and reaction orders from chemical kinetic data from various reactors (batch, continuous stirred tank, plug flow) and reaction networks. Additional capabilities of hierarchical error modeling and prior distribution specification make CKBIT a flexible, accurate tool for the task of kinetic parameter estimation and uncertainty quantification.

petBOA is a Python-based Parameter Estimation Tool utilizing Bayesian Optimization for gradient-free parameter estimation of expensive black-box kinetic models. We provide examples for Python macrokinetic and microkinetic modeling (MKM) tools via a wrapper interface, such as OpenMKM. petBOA leverages surrogate Gaussian processes to approximate and minimize the objective function designed for parameter estimation. Bayesian Optimization (BO) is implemented using the open-source NEXTorch/BoTorch toolkit. petBOA employs local and global sensitivity analyses to identify important parameters optimized against experimental data, and leverages pMuTT for consistent kinetic and thermodynamic parameters during estimation. petBOA also supports Docker SDK for Python to support containerized black-box models, such as OpenMKM

Artificial Intelligence Molecular Similarity (AIMSim) is a cheminformatics platform for performing similarity operations on collections of molecules (molecular datasets). AIMSim provides sophisticated similarity operations, molecular fingerprinting, and multiprocessing capabilities. With thousands of available chemical descriptors and almost 50 unique metrics for calculating similarity, AIMSim provides a unified platform to simplify cheminformatics workflows, such as diversity quantification, outlier and novelty analysis, clustering, and inter-molecular comparisons.

Estimating the thermochemical properties of systems is important in many fields such as material science and catalysis. Particularly, heterogeneous catalytic kinetic models, where few experimentally derived thermochemical properties are available, rely on First-Principles (density functional theory) and Semi-Empirical First-Principles (group additivity, Bronsted-Evans-Polyani relationships) methods to estimate these properties. The Vlachos group’s Virtual Kinetics Lab offers two open-source Python packages instrumental in facilitating these approaches.

The Python Multiscale Thermochemistry Toolbox (pMuTT) implements statistical thermodynamics for thermochemistry calculations and allows a one-shop calculator. Conversion between quantum mechanically computed properties and thermodynamic properties is ubiquitous in multiscale modeling. pMuTT converts data from a) experimental observations or b) ab-initio data or DFT calculations to thermodynamic properties of species and reactions and kinetic parameters of reactions. Several input and output formats are enabled to make the tool useful. pMuTT offers extensive functionality right out of the box to automate these routine, and repetitive tasks. pMuTT also has a comprehensive documentation page with many clear examples.

As kinetic models grow the number of adsorbates and reactions grow making First-Principles methods prohibitively expensive. Semi-Empirical First-Principles group additivity offers a route to quickly parameterize a model’s thermochemistry speeding overall model development.

Python Group Additivity (pGrAdd) is a Python package and database that implements the First-Principles Semi-Empirical Group Additivity method for estimating thermodynamic properties of molecules in the gas phase, liquid phase, and on catalysts. pGrAdd allows researchers to rapidly estimate the thermodynamic properties of thousands of molecules/adsorbates and of large molecules from thermochemistry of a smaller dataset of small molecules, building and deploying models in a fraction of the time normally required. pGrAdd contains group contribution databases of gas species and adsorbates on Pt(111) and Ru(0001) surfaces so new models can be built immediately. In addition, a user can easily build a new database from their own DFT data for any adsorbates and surfaces they require.

VLab enables the multiscale modeling workflow by integrating first-principles calculations and data-driven methods to calculate thermodynamic properties of adsorbates on catalysts, reaction rate constants, kinetic models, reaction pathways and networks and their visualization and analysis. To understand and optimize reaction kinetics, VLab provides an open-source microkinetic modeling software, a post-processing tool for visualizing reaction flux, as well as a data management software for the entire multiscale modeling workflow..

Open-source Microkinetic Modeling (OpenMKM) is a toolkit for modeling multiscale homogeneous reactions, e.g., gas-phase, and/or heterogeneous catalytic reactions. Microkinetic modeling enables coupling of “microscale” atomistic data with “macroscale” reactor observables. OpenMKM is a modular, object-oriented, C++ software toolbox built upon the popular and robust open-source Cantera software. With OpenMKM, users can quickly set up and start performing microkinetic simulations, for different ideal reactors such as, batch, CSTR, PFR without the need to write any code. OpenMKM is tightly integrated with the pMuTT (input file generation) and ReNView (post-processing) software.

Reaction Network Viewer (ReNView) quickly generates a graphical representation of the reaction fluxes within the system essential for identifying dominant reaction pathways and reducing a mechanism without undergoing manual data processing. ReNView helps users analyze reaction mechanisms and identify key species and reactions by showing the flux for each path (via line thickness), partially equilibrated reactions (via line color), and surface coverage (via species background colors). Relative magnitude and the direction of each reaction flux is also portrayed for all the pathways.

Chemical Kinetics Database (CKineticsDB) is a data management framework and software for multiscale modeling in heterogeneous catalysis. It is extensible and adaptable, with a data model rooted in the inherent relations of the research workflow, resulting in efficient data management. CKineticsDB retains all the information from simulations at various scales and allows accurate regeneration of publication results. The stored data can be accessed based on software parameters, catalysis parameters, reactions, and publications. The data is curated before uploading to the database through a semi-automated process to ensure computational diligence in calculations and uniformity in the stored files. The CKineticsDB software can be downloaded and used to manage research data locally as well as connect with a remote data server.