PAULYMigration Models
Longer description coming soon.
Empowering Your Packaging & Safety Workflows
Designed for the pharmaceutical and food packaging industries, this toolbox provides the critical predictive data needed to assess chemical migration and support FDA or safety reportings.
- Worst-Case Scenario Modeling: Safely estimate maximum leaching concentrations. Our tool computes robust worst-case steady-state scenarios to ensure the highest safety margins for your packaging evaluations.
- Mechanistic & Direct Predictions: Choose the level of complexity you need. Predict partition coefficients directly, or use our two-step LSER approach for deep mechanistic transparency.
- Future-Ready Kinetics (Roadmap): While our current models focus on establishing thermodynamic equilibria and steady-state limits, our roadmap includes advanced kinetic models to describe time-dependent migration profiles in the near future.
- In Development: Toolbox D is currently under active development. Connect with our experts today to shape the features and be among the first to deploy our leaching models.
Advanced Modeling for Polymer Migration
To calculate chemical leaching from polymers, we leverage our foundational machine learning architectures to predict the governing equilibrium states. You can model the migration using the following methodologies:
- Direct GNN Predictions: Utilize our best-performing Graph Neural Networks (GNNs) to directly predict critical baseline parameters, such as the octanol-water partition coefficient (log Kow), straight from the SMILES string.
- AI-Enhanced LSER Framework: Predict complex polymer/water partition coefficients with mechanistic clarity. We use GNNs to predict the underlying solute descriptors, then apply established LSER phase parameters to accurately describe the equilibrium situation between the polymer and the contact medium.
- Steady-State Calculations: Automatically integrate your predicted partition coefficients into established equations to simulate worst-case steady-state leaching scenarios.
Scientific Transparency
Our foundational models for partitioning and descriptor prediction have been rigorously benchmarked and published in peer-reviewed scientific literature, ensuring the fundamental physics of our leaching models are built on solid ground.
Publications
- "Predicting partitioning from low density polyethylene to blood and adipose tissue by linear solvation energy relationship models", J. Biomed. Mater. Res. Part B, 2023. https://doi.org/10.1002/jbm.b.35304