Prediction of Wear Using Multiscale and Multiphysics Approaches

Completed research project

Together with partners, Fraunhofer IWM has developed a methodological framework that, for the first time, enables the calculation of wear and friction in rolling bearings and gears across multiple length scales—from atoms to the component level. This provides a physically sound prediction method for limiting friction and wear that can be applied to other materials-lubricant combinations, thereby reducing the need for empirical testing and shortening development times.

Project description

Market Needs and Challenges

To meet rising efficiency requirements and stricter regulations, low-viscosity lubricants are increasingly being used, and components are operating under higher loads. This reduces lubrication clearances in rolling bearings and gears, leading to increased mixed friction. To date, there has been no reliable, physically based method for predicting wear, which complicates the development and validation of new components. Existing wear models rely on experimentally determined wear coefficients that are hardly transferable to other systems.
 

Project Approach

The project developed a multiscale and multiphysics methodological framework. Wear tests on real test benches are combined with high-resolution surface analysis of the resulting tribological layers as well as with molecular dynamics simulations of lubricants and interfaces. Constitutive laws for viscosity, wall slip, and boundary friction derived from these are incorporated into TEHD wear simulations.
 

Contribution to Addressing the Challenges

For the first time, it has been demonstrated that limiting friction coefficients can be predicted atomistically without purely empirical calibration and agree well with experimental values. The developed methodological framework is transferable to other material-lubricant combinations and provides industrial companies with a tool to digitally estimate friction and wear as early as the development phase, reduce testing, and design components more precisely.

Fraunhofer IWM Work Packages:

Using molecular dynamics (MD) simulations, Fraunhofer IWM modeled the lubricant behavior (PAO4 with ZnDTP additive) under high pressure and shear and derived constitutive laws for viscosity (shear thinning), wall slip on surfaces, and a combined friction law. Based on surface analyses, atomic models of polyphosphate tribo-layers were created, and their mechanical properties as well as nanoscale wear processes (asperity collisions) were simulated. Contact mechanics — calculations using real topographies yielded local pressure and gap height distributions. The result is a validated friction law implemented in Python that predicts limiting coefficients of friction as a function of pressure, temperature, shear rate, and gap height.

Based on the project results, Fraunhofer IWM can offer industrial companies the following research and development services:

  • Molecular dynamics-based lubricant characterization: Prediction of viscosity and shear behavior of new lubricant formulations under pressure and temperature without the need for extensive measurement campaigns.
  • Atomistic prediction of boundary friction: Calculation of friction coefficients in narrow lubrication gaps as a digital building block for the design of friction-optimized components.
  • Atomic-level tribolayer modeling: Elucidation of the effect of additives (e.g., ZnDTP) on protective layers for targeted additive selection and optimization.
  • Contact mechanics simulation of real surface topographies: Determination of local pressure and gap height distributions to evaluate the influence of surface roughness on wear and friction.
  • Nanoscale wear mechanism analysis: Elucidation of material transfer and cold welding in roughness contacts to optimize material pairings.
  • Integration of friction laws into TEHD wear simulations: Combining atomistic insights with macroscopic component simulation for reliable service life predictions.

Funding information